AUSCRIPT PTY LTD
ABN 76 082 664 220

Level 4, 179 Queen St MELBOURNE Vic 3000
(GPO Box 1114J MELBOURNE Vic 3001)
DX 305 Melbourne Tel:(03) 9672-5608 Fax:(03) 9670-8883

TRANSCRIPT OF PROCEEDINGS


O/N VT10307

A3.4.03

AUSTRALIAN INDUSTRIAL

RELATIONS COMMISSION

JUSTICE GIUDICE
VICE PRESIDENT ROSS
VICE PRESIDENT LAWLER
SENIOR DEPUTY PRESIDENT WATSON
SENIOR DEPUTY PRESIDENT LACY
COMMISSIONER HOFFMAN
COMMISSIONER LARKIN

C2002/2281, 2282, 2283, 2284,
4268, 5545, 5546, 5547, 5548,
5558, 5559, 5569, 5639, 5640,
5674, 5679, 5692, 5693, 5694
and 5716

CHILD CARE INDUSTRY (AUSTRALIAN CAPITAL
TERRITORY) AWARD 1998

LAUNDRY INDUSTRY (VICTORIA) AWARD 1998

BUILDING SERVICES (VICTORIA) AWARD 1994

THE HOSPITALITY INDUSTRY - ACCOMMODATION,
HOTELS, RESORTS AND GAMING AWARD 1998

TIMBER AND ALLIED INDUSTRIES AWARD 1999

RUBBER, PLASTIC AND CABLE MAKING INDUSTRY
- GENERAL - AWARD 1998

GROCERY PRODUCTS MANUFACTURE - MANUFACTURING
GROCERS AWARD 1996

STORAGE SERVICES - GENERAL - AWARD 1999

COMMERCIAL SALES (VICTORIA) AWARD 1999

VICTORIAN LOCAL AUTHORITIES AWARD 2001

CLERICAL AND ADMINISTRATIVE EMPLOYEES
(VICTORIA) AWARD 1999

TRANSPORT WORKERS AWARD 1998

GRAPHIC ARTS - GENERAL - AWARD 2000

METAL, ENGINEERING AND ASSOCIATED
INDUSTRIES AWARD 1998

HEALTH AND ALLIED SERVICES - PRIVATE
SECTOR - VICTORIA CONSOLIDATED AWARD 1998

RETAIL AND WHOLESALE INDUSTRY - SHOP EMPLOYEES - AUSTRALIAN CAPITAL TERRITORY - AWARD 2000

VEHICLE INDUSTRY AWARD 2000

THE VEHICLE INDUSTRY - REPAIR, SERVICES
AND RETAIL - AWARD 1983

HORSE TRAINING INDUSTRY AWARD 1998

CLOTHING TRADES AWARD 1999

Applications under section 113 of the Act
and section 108 References to a Full Bench
by the applicant unions to vary the above
awards re Safety Net Review 2003

MELBOURNE

10.05 AM, WEDNESDAY, 2 APRIL 2003

Continued from 1.4.03

JUSTICE GIUDICE: Yes, Mr Harnath.

PN1121

MR C. HARNATH: Thank you, your Honour. If the Commission pleases, I just wish to amend my appearance, together with myself appearing for the Association, from today, is MR M. MOIR.

PN1122

JUSTICE GIUDICE: I am sorry?

PN1123

MR CAIN: Mr Moir.

PN1124

JUSTICE GIUDICE: Fine, good. Thank you. Dr Kates.

PN1125

DR KATES: Yes, thank you, your Honour. Yesterday we were going through the material in ACCI4, which dealt with the various aspects of what we see as evidence that the increases in the safety net last year did have effect son the Australian economy. We say that there were effects on private sector employment, which you can see in the hours work data, that it would have contributed to the deterioration in the current account. It would have kept inflation higher, and to some extent, but to an unknown extent, it affected and has affected investment expectations.

PN1126

Now, I would like to turn to what I consider a seriously grievous aspect of these proceedings, and that is the - what is to my way of thinking, what looks like a series break in the labour force statistic published by the ABS. It is a -the labour force is, of course, one of those crucial figures published that we have. It is in fact the most high cost publication produced by the Bureau. We are unique, I think, in the world, in having a labour force which comes out monthly. Most economies are quite content to look at these things on a quarterly basis. But here in Australia we do it on monthly basis, and we have been doing this since the 1970s. And it is a crucial indicator and everyone uses it to get some kind of sense of what is taking place. Because without an accurate estimate of what is going on in the labour market, I think in many respects we are flying blind in terms of economic policy.

PN1127

Our concern is, and it is a very deep concern, that there is some kind of a series break which has taken place in the labour force data, which has made it in fact an entirely unreliable indicator, at least in the short term, of what is taking place in the labour market. The concern we have is, of course, not just the way the labour market data will be perceived by the Commission in deciding in terms of the effects of last year's increases, but will also affect the Reserve Bank, for example, when they think the economy is performing extremely well, when in fact the data, if it were properly - properly measuring current circumstances which show an entirely different - or were given entirely different perspective.

PN1128

Now, If I could take the Commission to ACCI4, and to the next section, which deals with the labour force. And the point about the labour force data is that, in our view, any examination of it should make one extremely suspicious about what the data shows. The ABS itself is quite clear that there is some reason for doubt about the accuracy of these figures, and as we note in our submission, in our original written submission on pages 3-10 and 3-11, the Bureau made the following statement its January publication. And what it said was:

PN1129

Employment estimates for January 2003 may have been influenced by the following factors and hence should be used with caution.

PN1130

And then they go on and they said this:

PN1131

The ABS is in the process of introducing a new sampling for the labour force survey based on the 2001 Census of Population Housing. The new sample will be introduced progressively over the period November 2002 to June 2003. While some additional volatility can be expected during implementation, the part with the new sample introduced in January shows considerably higher employment levels, particularly for males employed full-time and for females employed part-time in the part of the sample it replaced.

PN1132

What is really important to recognise in that statement by the ABS is that there is a suspicion that hangs over the ABS data, and going beyond that, that there is a reason for why things might have gone wrong. And the reason for why things may have gone wrong is the introduction of the new Census Population Housing data into the labour force data, which when they do that, rather than the normal sample rotation where they sort of - when someone finishes, they go to the person next door, which is almost literally true. And they try to maintain almost the identical sample.

PN1133

What they have done is entirely discard, and each month has come up, one-eighth of the sample, and have put an entirely new one-eighth of the sample based on a just completely new estimation of who is working and where the population is found. And this is because of the new re-basing by the - from the census data. Now what is our concern, and I think it is a concern that is shared pretty well across the map, is that something has gone wrong in this changeover.

PN1134

Now what I want to discuss is the fact that there is reason to be suspicious, and very strong reason to think that now, as this time, as we sit in this case with the new data being introduced, that one cannot use the labour force stats to gauge what has been taking place in the labour market since the introduction of new census data beginning in November. And if we look at that, the first section there, the first table in that, in ACCI4, showing the participation rate, and what is show in the data is the seasonally adjusted figures from July 2001 through until the most recent figures we have, which is at February 2003.

PN1135

And what I wish to point out is that if there is between October 2002 and November 2002, there is a very clear evidence that something has happened to the data that is taking it into realms in which it is not accurately portraying the underlying reality of the economy. Now, in November, the increase was on the monthly, 0.8 per cent, which is not of itself an unusual figure. If you look up the table you will see that in January 2002, there was a 0.9 per cent. So these sorts of things, given the nature of volatility in the statistics do occur and have occurred in the past.

PN1136

But what - if you look at the January 2002 figure, where it was 0.9, an exceptionally high figure, it was immediately followed by three straight negative figures which took the growth rate back down to what is the more typical experience that one expects at the - in the growth in participation in the labour force. Going to October, itself, you will see that in October 2002, the monthly increase was virtually nothing. And the percentage increase across the year to that date, had been 0.9 per cent. And that was - that has been the experience in the Australian economy for a considerable period of time.

PN1137

So if you go back it is that kind of growth rate, in participation, has been what we have normally expected to take place and given the variations that do happen month to month, that is the kind of figure that we would have typically expected. But as soon as we move into November, there is a very different aspect to the data and it grows by 0.8 per cent in November, which is not unusual but also, in our view, high by these standards. And you would expect, after that, that the numbers would come back and recede.

[10.14am]

PN1138

In December, however, there was another increase and this one was a 0.6 and then again there was an increase of 0.9 in January. So you had, over those three periods, you had an increase of 2.3 per cent in just those three months. And finally, in February, there has been something of a settle back where it is minus 0.2 per cent in the month.

PN1139

And it is entirely possible that we are looking at something that, in the longer run, will just simply wash up. So that you come back a year from now and you would find that this was an aberration and the - as a sample, was expanded into the new census data, you would find that these figures will have settled back into the pack and the figure would then again become 1 per cent, 1.2 per cent, 0.9 per cent, the kind of annual growth rate. But as we show at the bottom, here, we are looking - - - 

PN1140

JUSTICE GIUDICE: Excuse me, Dr Kates, could you just remind me where in the material the ABS bulletin is that deals with this issue of the reliability or otherwise of the - - - 

PN1141

DR KATES: It was, in fact, quite unusual. In the January publication they actually had a note on the front cover that said, see page 2.

PN1142

JUSTICE GIUDICE: Yes, but I am just wondering where it is in the material before us. Could you at some stage tell us where it is?

PN1143

DR KATES: Yes, certainly, I am happy to do that.

PN1144

JUSTICE GIUDICE: I think it has been referred to a couple of times.

PN1145

DR KATES: But what we are looking at here is a figure where, if you go from February 2002 to October 2002, the increase in the participation rate - on labour force participation, that is the number of people actually entering the labour market, either as employed or unemployed, has increased by 0.4 per cent and that is across that eight month period. But then you go to complete the year, the figures from October to February, and you find an increase of 2.1 per cent. An extraordinary large increase, even if it were across the year, never mind for a single four month period, and why this is a matter of genuine concern to us is that the participation rate, the labour force participation, will divide itself into basically the three categories.

PN1146

You will either show up in the figures for full time employment, part time employment or in the unemployed. And if you have a growth in the estimation in the figure, published figure, of participation, which in fact is unrelated to the actual circumstance of the economy, you will end up with and as I believe we have ended up with, a figure of - you will have a greater figure showing larger growth in full time employment, part time employment and even in unemployment than is actually warranted by the underlying economy. And it is not actually the point that the number of persons participating may actually be more accurate than the one that was there before the census data was rolled in.

PN1147

The important point is to recognise that if there is a series break between the October and November months, then what you have is, at this continuity between the growth rates, you can't actually look at the figures showing the growth between those figures - those periods, and feel that you seeing what actually transpired out in the economy. Because were there suddenly a mad rush of 2.1 per cent addition into the labour market, between October 2002 and February 2003, one would presumably have seen some evidence of that somewhere else in some other statistic, in some other aspect of the economy as we now are observing it.

PN1148

Now, if I could then take you to the next stage. And this is from our written reply submission, at page R1-9. And this shows the data on full time employment which are the figures we have continuously used ourselves as the measure that really does get some kind of sense of what is taking place in the labour market. We certainly agree that the flexibility provided by part time work is enormously important, but as a measure of the strength of the labour market the one figure we think really does capture mostly closely what is important is what has been happening to full time employment. And we have said that over a course of many years in these submissions.

PN1149

And what we find here is that there is a very clear division between the period up until October 2002 and the period since. That, if you go over the year to October 2002, the growth rate in full time employment was 0.8 per cent and in the month of October, itself, it was minus 0.5 per cent. Now, in our view, that is the story of what the labour market is about, that is what has been happening. This kind of not good, not bad but basically very sub-optimal improvement in full time employment, so that there has been growth but it has not been the kind of robust growth that you would actually like to achieve. You would really like to have a figure something like the 2.8 per cent that is recorded in the year to February 2003, but what you find, before the series break, is that the growth rate has been 0.8 per cent.

PN1150

Then you go across the divide between the old labour force data and the rolling in of the new figures, based on the new census material, and you find that in November the increase is 0.7 per cent, in December it is again 0.7 per cent, in January 2003 it is 0.9 per cent and then in February it has settled back and it is 0.1 per cent. So what you have across this divide is a new world of employment growth that is not corroborated by any other figure anywhere else. If you look at the ANZ job ad series which the ACTU quoted, you will find that their annual figure has been negative. If you look at the Department of Employment and Workplace Relations leading indicator series, you find that it has been falling for the past five months.

PN1151

So that there is no corroboration, there is nothing to make you think that anything has happened other than that there has been a series break in the figures published by the ABS. Now it is clear that - I can quite understand why no government would start a witch hunt when they saw figures like this because I think the Government was quite content to have the figures show 2.8 per cent growth. I also think, and I actually share this view, that we are not interested in seeing ABS data discredited. So that what I say in this Court room I do not take with me outside, I don't go around publicising this or making a big deal about it.

PN1152

[10.25am]

PN1153

I think in terms of what happens in here it is crucial that we appreciate that something has gone wrong in the ABS data and it is for that reason that the ABS is putting up warning signs and cautionary notes and suggesting and recognising that there is something that is suspect about these data. And if you look at the bottom of this page and this is a table taken from page R1-10 of our - or 1-10 of a reply submission, what you see there is that - again you break it down into two periods and you get the period from February 2002 to February - to October 2002 and the growth rate is 0.3 per cent.

PN1154

That is a really poor rate of growth, but I think it is more or less the accurate one. It suggests that the increase across the year had it continued at that rate might have been half a per cent, maybe a bit better.

PN1155

JUSTICE GIUDICE: Dr Kates, assuming we are all interested in sums and objective truth about this, and have objective truth - subjective truth, but presumably the ABS decided to alter the basis of its sampling because of some perceived deficiency in - or short coming in the sample it was using. It wouldn't have - do you know why it was altered?

PN1156

DR KATES: Yes, your Honour. They do this every five years, that every five years the new census date is rolled in and one of the reasons that they - one of the most important uses of census date is to rebase the labour force data so we know from the census the employment figures and then we - then they are incorporated into the new labour force sample.

PN1157

JUSTICE GIUDICE: But doesn't that suggest, and I am not doing anything more than suggesting it, but doesn't that suggest there might have been something wrong with the base that was being used?

PN1158

DR KATES: Yes, I agree the - what happens is clearly that every five years, by the time you have got to the end of it, the - - - 

PN1159

JUSTICE GIUDICE: Yes, the base is out of date.

PN1160

DR KATES: The base is clearly out of date so that it is necessary to actually do this change over, if you like.

PN1161

JUSTICE GIUDICE: Yes, yes.

PN1162

DR KATES: But - and well, what seems to have been the case in the past that there is nothing that shows up, you have just seen this change and they do it one month - each month they just roll on a bit more sample.

PN1163

JUSTICE GIUDICE: Yes, it is phased.

PN1164

DR KATES: That is right.

PN1165

JUSTICE GIUDICE: Yes.

PN1166

SENIOR DEPUTY PRESIDENT WATSON: Yes, Dr - I am sorry, had you finished? Dr Kates, how are we to balance your fairly impressionistic assessment of the change in the data to the comments of the ABS in the February labour force publication, which I think Mr Watson tendered, would suggest that they undertook some statistical analysis and concluded that the change had not appeared to affect the figures?

PN1167

DR KATES: Well, I think that the ABS is not of a mind to draw a line in its labour force data that says there has been a series break here. I think they will resist that to the final - to the crack of doom.

PN1168

SENIOR DEPUTY PRESIDENT WATSON: You are not suggesting the ABS is fudging on - - - 

PN1169

DR KATES: No, I am saying that a series break, if it has occurred, we will not know about it for a while, and in the meantime they are investigating for the same reason that we are suspicious, they are suspicious. There is such overpowering reason to think that these numbers are telling a - or giving a misleading picture of what is taking place in the labour market that there is now a review going on in the ABS and I am going to be part of a meeting next Monday in which the ABS will again examine these figures.

PN1170

So it is a matter of genuine concern to the ABS and they are doing what they can to try to see what has happened, and when push comes to shove it is not what I would say an impressionistic, it is one where I think you could take logical inferences that if you have sudden change in the direction of a measure and you know that there is some underlying reason that that may have taken place, which in this case is the rolling in of the census data, then I think the proper conclusion to come to is to say there well may be something, some break in that series that makes the most recent data an inaccurate measure of what has actually taken place.

PN1171

SENIOR DEPUTY PRESIDENT WATSON: But aren't you asking us to place greater weight on your inferences than the comments of the statistician, the official statistician, based on some actual analysis of the material?

PN1172

DR KATES: Well, I - well, as we have been talking about - with - in regard to our own surveys there are - even the best surveys can have a period in which the sample is simply wrong, that no matter what - no matter what safeguards you may put in in advance, in the end of the day, if the numbers are wrong because you happened to have picked a sample that is completely different from the actual population, and that can happen in even normal circumstances. What we are saying here is that this isn't even a normal circumstance.

PN1173

We are talking about the actual changing in the actual sample base, the normal system where you try to match the person moving out with the person moving in so that there is no discontinuity has not been done. There is a complete rebasing of that sample, total holus-bolus, so that a whole one eighth goes out and a wholly, entirely different one eighth goes in, and what we are saying here is that it is so obvious that between October 2002 and February 2003 that there wasn't an increase in full-time employment of 2.5 per cent, that the entire use of the labour force growth figures has got to be put into the suspect category.

PN1174

You cannot accept these numbers. Nothing else that has occurred in the economy beyond those data would corroborate those figures. There is nothing, no other measure that will tell you that the growth in full-time employment suddenly, out of nowhere, grew by 2.5 per cent, when in the year all the way up to October 2002 the increase was .8. You then go another four months and it increased by 2.5 per cent, and this is in a period of such extraordinarily high uncertainty where the ANZ job series is showing some weakness, where the Department of Employment and Workplace Relations figures are showing a negative figure, where their professional employment vacancy levels are showing a fall.

PN1175

All these things are occurring and yet you have, in the very month after the new census data is brought in, a sudden explosion in employment in the data which has no corroborating circumstances outside this courtroom to support it.

PN1176

SENIOR DEPUTY PRESIDENT WATSON: Well, it has the normal sorts of tests that the statistician does with the changes of series, when he changes the sample and it is quite a regular occurrence in a range of series and a normal process of then applying tests to assess the effect of those changes, so you have that don't you?

PN1177

DR KATES: Well, I don't know what tests they do. They actually just - - - 

PN1178

SENIOR DEPUTY PRESIDENT WATSON: Well, they are described in the February labour force report.

PN1179

DR KATES: Well, they actually say that there is a - that if they use match samples that you end up with 30,000, but I would even want to know whether they are matching the sample with October or whether the matched sample for February happened to be the matched sample with January, because once you are already into January you are looking at a polluted data sample. From November on data set is wrong and in our view it is clearly wrong and impossible to believe.

PN1180

You then go to December and January and if they then go into February and say, "Well, look we are just going to do a match sample in February with what we had in January", then of course they are going to get this kind of outcome, but it is already too late if you ask me.
[10.35am]

PN1181

Now, this is one of the things that I am going to find out when we meet with the ABS on Monday, and this is something which the ABS itself has recognised, and they have recognised it by the very virtue of the fact that they mentioned it in their publication, and this is the single most expensive survey that the ABS runs, and if this one is going off the rails, that is a serious blemish.

PN1182

I, like they, hope that if you keep going another few months, the thing will settle down and become comprehensible within the entire fabric of the series. But at this stage, as it shows at the bottom, it is almost - well, I say it is impossible to believe that between February and October 2002 you had growth of full-time employment of 0.3 per cent, and then in the succeeding four months you had growth of 2.5 per cent. I would say that that is improbable to the point of impossible. It is just, in our view, not possible that that has happened.

PN1183

We divide the data there, as you see, again into the period from December 2001 to May 2002 and we picked the five months up to the decision and that was 1.1 per cent, and then we picked the five months following the decision, that is a minus 0.4 per cent. But even if you take the longer period so that - because the May number itself was extraordinarily high and turned out to be a false dawn, going from October 2001 to April 2002, which is a six-month period, was 0.3 per cent. April 2002 to October 2002 was 0.5 per cent. So in each of these six months periods you have an increase of around 0.5 per cent. Suddenly, in the next four months, immediately after the introduction of Census data, you have a figure of 2.5 per cent.

PN1184

I would say, very strongly, that that 2.5 per cent is suspect to the highest degree. You cannot, I think, accept that a figure like that at a time like this in an economy as we now have, that is an invalid number, and we would find it - - - 

PN1185

JUSTICE GIUDICE: Dr Kates, I think you have made the point several times. I am not saying that you haven't been provoked into repeating it by me and - - - 

PN1186

MR KATES: No, okay. Well, I will go on.

PN1187

JUSTICE GIUDICE: But I think we have got the point you make. At the end of the day, it is a question of judgment, to some extent, as to whether one accepts the basis upon which you say that the figure is wildly improbably or highly improbable.

PN1188

MR KATES: Thank you, your Honour.

PN1189

SENIOR DEPUTY PRESIDENT WATSON: On a more general issue in relation to employment, isn't one of the issues of the last eight years or so that Australia has had very good growth relatively and absolutely, but it has equally had extremely good productivity performance so the growth has not flowed into employment to the degree that one might have hoped for the sake of employment. And that raises a dilemma, does it not, of, in the short term at least, trade-off between productivity and employment, understanding in the longer term productivity is important for future employment growth, but isn't that really a story of the last decade or so that the growth has really been reflected in productivity so we haven't seen as good an employment outcome as we might have hoped?

PN1190

MR KATES: Yes, I think that is exactly what has happened, and that is one of the issues we tried to raise in these proceedings, that for one reason or another, the forms the growth has taken have tended to be in ways that minimise additions to employment. They have occurred, of course, but they have been in areas which are either highly capital intensive, or the have been done in ways that minimise the additions of new employees, or they have been undertaken in ways that substitute capital for labour.

PN1191

One of the things that we argue is that this improvement in productivity is partly a reluctance on behalf of business to employ additional employees. Now, one of the reasons for that, I think, has been the increases in the cost of employment, and it is not just to the safety net, of course. I mean, the Superannuation Guarantee, I think, has been an absolute disaster for employment, but these are the kinds of imposts that have been laden on to businesses that have made them less likely to employ, and I think these are genuine problems.

PN1192

If I could now turn to looking at the Australian economy and at the economic data that we have provided. This is from the next section of our - ACCI4 and this provides the review of the economic statistics. If I could just go through these data and discuss them. Now, the general story, I think, that I would tell in terms of these figures is that when you look at them, you can see a softening in the economy, and it becomes unmistakable.

PN1193

If you just start with the very first table which is on Gross Domestic Product, and you can see both in the data sets where at September 2001 the quarterly growth rate was 1.3 per cent and by December 2002 it had fallen to 0.7 per cent, and the annual growth rate up from March 2002 has gone from 4.5 per cent to the most recent figure of 3.2 per cent. This is reflected in the chart which shows a growth up until September 2001 and then since then a pretty well continuous fall off in the trend level of quarterly growth in GDP.

PN1194

Now, if I could turn to the second table which is on market sector GDP. As we have said before, this is our measure of the private sector, the closest thing you can actually get to measure the private sector, but by being the market sector it is the one area where goods and services are actually sold at a price and profitability is measured so that you can actually calculate the growth in an accurate way that actually gives you a sense of value added. Otherwise, it is simply employment growth that is used as the proxy.

PN1195

What you see in this figure, which is the private sector proxy, is that in the quarterly figure, again there is a maximum reached at September 2001, and since that time there has been a downwards trend in this figure so that as at December 2002 it had fallen to 0.7 per cent. This is shown in the chart as well. If you look in terms of the growth in the annual figure, it is 5.4 per cent at March 2002 and it has fallen by December to 3.9 per cent, and given the nature of this figure it is very likely to be falling further. Looking at the December quarterly figure of 0.7, that is an annualised growth rate of 2.8 per cent, and I think it is entirely possible that the growth rate will fall substantially below the 3.9 per cent.

[10.45am]

PN1196

We were talking about what is happening in the private sector. That is the closest measure we can come to in an aggregate sense and it is a pretty grim story. Looking at table 3, which is non-farm gross domestic product, the non-farm is the GDP which removes the agricultural sector and in this case it includes a very large measure of public sector in that. So that given the expenditure that has been taking place in terms of national security and then war between Iraq - and East Timor - plus other forms of public sector spending, we see an appearance there of a continuing and ongoing growth rate, but as the figures on market sector growth show, that it is being held up to a very large extent by public sector spending.

PN1197

If you think in terms of what is happening in business that is a very different kind of story entirely. Looking at table 4 the figure there is the sorrow, the pity. This is an extraordinary, an extraordinary .....

PN1198

THE SENIOR DEPUTY PRESIDENT: Dr Kates, why isn't the difference between chart 2 and chart 3 explained by the graph?

PN1199

DR KATES: Pardon me?

PN1200

THE SENIOR DEPUTY PRESIDENT: Why is it explained by the factors that you have asserted just then? The declining growth - rate of growth in chart 2 since June or September of 2002, and the more satisfactory position shown in chart 3, you sought to assert that the more satisfactory position in chart 3 was explained by the level of public spending - you mentioned East Timor - and the public sector generally. Why isn't the drought the explanation?

PN1201

DR KATES: I would say, your Honour, it is a very large part of the explanation. It is - the drought has taken an enormous toll on Australian economic growth and certainly both in GDP and in the market sector. What you are looking - - - 

PN1202

THE SENIOR DEPUTY PRESIDENT: So if the drought is breaking, and what is occurring at the southern oscillation index suggests that the El Nino is well and truly finished - if the drought is breaking why can't we expect a rebound in the - or a change in the trend - in the downward trend in chart 2?

PN1203

DR KATES: Well, I would - our own view is that this will trend down a bit and when the drought does start to break and it shows up in the data stats, you will find that that will give an upward impetus to these figures. But at the same time as that is happening, looking at the domestic economy and into the non-farm sector, you will find that the growth rate in private consumption is likely to fall, which is what most forecasters are showing. There is likely to be slower growth in business investment and there is likely to be a contraction - an actual contraction in investment in dwellings.

PN1204

So that we will get this turnaround hopefully in farm sector product but it will be more than overcome by the effects on the non-farm sector which are deteriorating - not contracting, but in most aspects will show much slower growth in the year ahead than has occurred over the past year. Now, the data on farm sector shows this extraordinary fall of 20.9 per cent over the past year in gross farm value added, and what is really important about that is that is translating across the economy, not just into those particular farms, but is affecting businesses in the rural sector and is affecting businesses more remotely, but certainly across the Australian economy.

PN1205

This fall off in addition will make it less - will make the economy less capable of sustaining an increase in the cost of labour. It will certainly make it, in our view, impossible to sustain an increase in the cost of labour in the rural sector. It will affect related industries such as agricultural machinery very badly, so that we are looking at what may well be a temporary circumstance, although we do not know that yet, but is one that has to be factored into any decision that is made. You cannot have the ACTU saying, well, the drought hasn't happened yet and then come and then say, well, it is about to end, so at what stage does one actually say, well, that growth has affected the Australian economy and, therefore, should ameliorate the decision in the safety net.

PN1206

Our view is that this growth, which has not yet ended, and whose effects are certainly going to last more than until the next rains arrive, is a circumstance of great importance in deciding what ought to be determined in this case. Turn to private consumption expenditure, which is table 5, and again you see that there is this falling off of private consumption expenditure. You can see the growth rate was at its peak in March 2002 when a group - in the quarter by 1.3 per cent and since then it has fallen in each of the succeeding quarters and at December 2002 it was down to 0.7 per cent.

PN1207

I would say that this is very distantly related to the drought but would undoubtedly to some extent have - and affected by the drought, but is more likely to be just a general circumstance of the economy and I would relate it more to the fact that the savings ratio in Australia - the level of savings to household income is actually negative in the last year - in the last quarter and there is - to the extent that consumption has been maintained it has been maintained by people actually digging into their savings. If we turn to table 6, which is private growth fixed capital, and this has been, no questions asked, a very good year for investment in the private sector.

PN1208

The growth over the past year has been 17.0 per cent. A very great deal of it, and very important factor, has been the First Home Owners Scheme which affected dwelling investment, but very importantly, this has taken - this has occurred over the past year and it has been part of the reason that the Australian economy has remained so buoyant. Our concern is that this is not going to be the case next year. The 17 per cent will just simply not eventuate.

PN1209

We may have - we will have a very low figure a year from now - that at the very beginning of that will be the negative growth expected by most forecasters in dwelling investment, which I will now turn to the table 7 which shows the figures in dwelling investment.

PN1210

[10.53am]

PN1211

This has been an amazingly strong and resilient recovery in dwellings. The quarterly growth rates are - some of those quarterly growth rates could have been annual growth rates and you would not have complained. The figure even at December 2002 has been 4.8 per cent, but as an overwhelming preponderance of data shows, which we will come to, the - - - 

PN1212

JUSTICE GIUDICE: I am sorry, Dr Kates, what was that? An overwhelming preponderance of what?

PN1213

DR KATES: Data. The growth in the dwelling construction area is about to come to an end. The cycle has reached its peak, and is coming down, and is likely to come down with a very, very large descent in the growth rate between what you see here, and what you will see a year from now. The First Home Owners Scheme has come to an end. Housing starts are slow, and the data on loans for new housing is coming back. So that while there is at this stage a very good rate of growth, 22.7 per cent, it is unlikely to be sustained by the bushfires in Canberra.

PN1214

If I could then turn to non-dwelling investment, Table 8. The growth rate across - through to 36.8 per cent, is quite spectacular, and a very important part of what has made this year as buoyant as it has been. Again, as the business investment data from the private capital expenditure series has shown, the kinds of growth you see in this table are unlikely to be repeated in the year ahead.

PN1215

Similar, if you look at Table 9, which is machinery and equipment investment, you see again a very good year, and 11.4 per cent and the growth rate still showing residual strength even into the December quarter. The two last tables are summarised in Table 10, which is the private business investment series. This is, in essence, the non-dwelling investment, and machinery and equipment investment.

PN1216

Over the past year the growth in private business investment has been 16.6 per cent, and while you could say that there has been some slowing in the trend growth rate in the last quarter, I would say that it is still showing good growth all the way into the December quarter, even if it has to some extent levelled out. If we could keep that 3.9 per cent each quarter forever, I would be very, very happy. But as Table 11 shows, that is very unlikely to be the case.

PN1217

This is a very complex table. If I could just take you to Estimate 5, which is the third one from the right. What Estimate 5 is made up of, is six months actual for the current financial year, so it includes the September and December quarters of 2002: and then it adds on the expectations going forward for March and June. So that what you see there is the figure of 15.2 per cent

PN1218

So that taking the two halves together, the actuals for the first half of the financial year and the expectation for the second half of the financial year, then you have a growth rate for this financial year of something like 15.2 per cent. And that is, of course, reflected in the data as we now have them.

PN1219

But if we move to the first column on the left, what you see there is the first estimate for 2003/2004. What this shows is that in January and February just past, the ABS does an estimate of what business expects to occur in the financial year 2003/2004. So basically this forecast occurs something like six months before the financial year begins. What we have here is a figure that is substantially lower than the figure of the previous year, which was 14.4 per cent.

PN1220

I would also remind the Commission that when we were in submissions last year, the figure that was published by the ABS at that time, and the most recent one we had available, was 21.2 per cent. So that even the 4.5 per cent is still a preliminary figure, so the effect on uncertainty of the current situation may yet make that number even lower.

PN1221

But the point that really has to be established is that the growth rate over the year expected for the year 2003/2004 is well down on the growth rate that we have just been through, that we cannot assume we are going to have a growth rate in private investment in the coming year as strong and as vibrant as the one we have had. And this figure does not even include the figures on dwelling investment: and since dwelling investment is very likely to come off well below the 22 per cent, and most forecast thinking has even been negative, the - - - 

PN1222

JUSTICE GIUDICE: Dr Kates, I may be stating the obvious, or I may be making a grievous error, but there does not appear to be a lot of correlation, or a high degree of correlation between Estimate 1 and Estimate 7, except in 2001/2, I think.

[11.01am]

PN1223

DR KATES: Yes, I agree. In fact the tendency has been that the first estimate is higher than the eventual outcome, so that what you see, for example, as a 17.3 in 1998/99 ended up as a minus 1.8 and - - - 

PN1224

JUSTICE GIUDICE: But it was the reverse in the following year.

PN1225

DR KATES: There was the reverse. It was a lower negative, that is true, so they were expecting a minus 14 and got only a minus 2, but the tendency has, to some extent, been that the final outcome is lower. But even if it is true that we are only looking at a kind of rough guess of what will happen in 2003/2004, this is, as it were, the best rough guess we have and the growth rate that we can expect for investment, as a kind of best guess, is about 4 and a half per cent, given what we now know, and as I showed in the age economic forecast the average growth rate was something like 7.8 per cent amongst all the forecasters there.

PN1226

So that this 4.5 and that 7.8, there is not a lot of difference between them and the point that we are trying to establish is that, on what we know right now, the growth in investment is likely to be lower in the financial year ahead than it has been in the one just past. Turning to table 12 which is on exports and imports, again I would agree that these figures are to a very important extent growth related, and they show the fall in the level of exports in the December quarter of 0.4 per cent and they show a very minimal growth rate of 0.4 per cent in the year to December. You compare those with the figures for imports and imports have been extremely strong, increasing by 13.8 per cent across the last year so that - - - 

PN1227

SENIOR DEPUTY PRESIDENT WATSON: Does that in any way reflect the high levels of investment as finished?

PN1228

DR KATES: Yes, it is. Yes, I agree that it is. It is certainly related to the level of investment and in fact that I think is one of the concerns, that what you have is an investment figure which shows installation of fixed capital in Australia, and then you look at the imports figure and you realise that even that investment expenditure that you had in Australia has largely been imported so that the investment we have, which will of course make the Australian economy more productive, has nevertheless not been, by and large, conducted inside Australia itself.

PN1229

So the archetypal example is, I suspect, the importing of Qantas airplanes so that they were produced in Washington State and they were flown into Australia, so we have a large increase on the import and a large increase investment, but no actual employment creating activity inside Australia itself. But I certainly do agree that it has been investment related. We now turn to the employment data, and again we really don't know how to deal with these figures other than to say you have a growth rate up until the September quarter in employment of 0.4 per cent in the quarter and 1.9 per cent across the year to September.

PN1230

You then have what are extraordinary and we say literally unbelievable figures for December and March where you have increases of 1.1 and 1.3 per cent, and the annual growth rate suddenly blows out to 3.1 per cent in the year to March 2002. We note that these are the preliminary figures because we only have the data up until February, but even doing that we just compare the three months to February with the three months to November, so that we try to get an accurate estimate of three months to three months.

PN1231

But what you see in the chart as well is this spectacular growth in employment, but it is one we would say that is simply a statistical error more than anything reflecting what has taken place in the economy. Table 15 shows full-time employment and here again you see the same story, and in fact the same story in a much worse fashion, that the growth in full-time employment was quite subdued over the year until September. In the year to September it had grown by 1.0 per cent and as we will show in the next table the growth in employment had largely been fuelled by part-time employment.

PN1232

Then in December you have a growth rate of .9 per cent and in March you have a growth rate even larger of 1.2 per cent, so that you have, as the chart shows, this spectacular increase in full-time employment that occurs for no reason we can actually think of in terms of what may or may not have been going on in the economy overall. That is why we think of the figures up until September as being extremely significant because we were not creating the full-time jobs we would like and the full-time labour market had been subdued.
[11.07am]

PN1233

A complete contrast to the part-time figures in table 16 where it will be seen that there was an increase in part-time employment of 1.0 per cent, that is in the quarter, and the quarterly increase in part-time employment to September is the same as the annual increase in full-time employment to September, so that there had been this huge imbalance in where jobs were being created, so that over the year to September the aggregate growth in part-time employment was 4.3 per cent compared with the 1.0 per cent in full-time.

PN1234

We then turn to table 17, and that shows unemployment, and what is really quite extraordinary when you think about it, is that shows that even with massive growth in employment and massive growth in full-time employment and massive growth in part-time employment you still have an increase in the unemployment level. Now, this is because, we say, this is because there has been something that happened to the data that just simply recorded the number of persons in the labour force that aren't actually representative of the true increase that took place, but had spilled over into all the categories of employment that are recorded in the labour force data.

PN1235

Up until September, however, and that is where I think the numbers are still acceptable, the figure does show a fall in unemployment of minus 1.6 per cent, but we also note that even that had been diminishing, although I wouldn't even then make much of it because of the nature of trend estimates. The unemployment rate is shown in table 18. The fall in the unemployment rate has been a very important feature of this economy. It has gone from 6.9 per cent in September 2001 to 6.2 per cent in September 2002.

PN1236

The continuous fall in the rate in December in March we would not see as - see in the same light as we do in the growth rates, since these are rates within the same month, so that it may well be an accurate picture of the unemployment rate itself, but as we noted in our previous tables, this growth in the - this improvement in employment has been largely and almost entirely occurring in that of the part-time employment.

PN1237

JUSTICE GIUDICE: They are seasonally adjusted, aren't they?

PN1238

DR KATES: No, these are trended, your Honour. It says, I think it is, yes, it says at the top. In fact, all of the data in this section is trended except for, I think, the wage cost index. The next table is that of the male full-time employment. It is one of the things that the labour market data show is that where the increase may have even taken place - the unacceptable increase may have even taken place is in male full-time employment. What you see here is that up until September 2002, which is what we say is the last date which you could actually take these figures at face value, the growth in male full-time employment had actually been negative, and the data show an actual contraction in the level of male full-time employment at that stage.

PN1239

This of course has been succeeded by two extraordinary months of - quarters where it is .6 in December and 1.1 in March, a 1.7 increase in just half a year in comparison with the 1.2 per cent growth rate that is recorded in the year to September. I now turn to table 20 which shows the consumer price index and we note how we are flirting with danger at every stage since March 2002, the Reserve Bank target range for an acceptable inflation rate is between 2 and 3 per cent, and since that date we have been kind of a range of ourselves between 2.5 per cent which is the low point in September 2001 and it is on three occasions crossed above the RBA target range and most recently in September 2002 where it went to 3.2 per cent.

PN1240

We now turn to the wage cost index. We have referred to this table earlier. It is the data set that goes into the chart is the private sector data which you see on the left. The point we have tried to establish using these figures is that when the safety net comes in in the September quarter you see it, it is visible, so that you can see the spike in each year in the data so that in September '98 it is 1.1 whereas the two surrounding quarters are 0.5 per cent.

PN1241

You go to September 1999 and it is 1.0 per cent for that quarter and the two surrounding quarters average out to 0.6 per cent. In September 2000 it is 1.3 per cent. The two surrounding quarters show an increase of 0.7 per cent again. The figure for September 2001 is 1.2 per cent, and the surrounding quarters are 0.6 per cent and in September 2002 it is 1.4 per cent and the two surrounding quarters are 0.7 per cent. And what we seek to demonstrate by this is that the decisions that occur as part of the safety net are not the trivial 0.1 per cent increases to labour cost, but are actual genuine increases that are entirely visible and palpable in the data set and I can assure you that they are identical in the actual experience of businesses.

PN1242

[11.15am]

PN1243

Table 22 is the real wage cost index which is the same data as we have had in the previous table except it has been deflated by the GST adjusted CPI, and you can see that across the year there has been a growth in the real level of the wage cost index of 0.4 per cent, and over the entire period of this table with the index at 104.6, the increase in the real level of the wage cost index is 4.6 per cent.

PN1244

Turning then to table 23 which is ordinary time earnings, and these figures are an important contrast to the ones in the wage cost index. This is what people get paid on average in the Australian economy and it is typically, but not invariably, higher than that for the wage cost index because it is not just through increases in wages for doing the same job that individuals get increases in their actual levels of payments. People are promoted within the workplace, people receive over-award payments, they move to different jobs, so that what these figures show are the kinds of increases that take place on average to employees across Australia, and in the most recent data the increase has been 1.1 per cent for the quarter and 4.7 per cent across the year.

PN1245

The next table, table 24, shows the figures for real full time adult ordinary time earnings, so these again have been adjusted by the CPI, and that is in the third column, real - the AWOTE, and what you can see again is that across the year there has been an increase in this measure of 1.6 per cent and quarterly increase of 0.4 per cent.

PN1246

SENIOR DEPUTY PRESIDENT WATSON: Dr Kates, I think you said the ordinary time earnings included overtime; is that what you said earlier?

PN1247

MR KATES: No, ordinary time earnings is ordinary time earnings. It does not include - - - 

PN1248

SENIOR DEPUTY PRESIDENT WATSON: Yes. But I thought you said one of the factors within it was - over-award, was it, over-award payments?

PN1249

MR KATES: Yes, that is right, but not overtime, no.

PN1250

SENIOR DEPUTY PRESIDENT WATSON: I am sorry, I misheard you.

PN1251

MR KATES: Table 25 is a very important table in our view. These are the data on GDP per hour worked, and this is across the entire economy, and what you see here is the GDP as an index and then you have GDP per hour worked as an index and GDP per hour shown in the third column. The growth in the year to December, which is the most recent figure, is 1.5 per cent. These are the total economy figures and they include the public sector and the non-market sector.

PN1252

We therefore turn to look at the market sector GDP per hour worked, and the data in that second column are data we have already presented to the Commission, and they show the growth rate in hours worked which has been quite static over that last three years. And in fact unusually for a chart, we have included the hours worked chart rather than that for the productivity itself because we think that it is the hours worked in the market sector which is the significant aspect of these data at this time.

PN1253

The growth in productivity across the year has been 3.8 per cent in the market sector which is made up of a growth rate of 3.9 per cent in GDP in the market sector which is discounted by the very minimal 0.2 per cent growth in employment and hours worked in the market sector, so that you get a productivity growth rate of 3.8 per cent.

PN1254

Now, I turn to deal with another aspect, and this is taken from our original written submission, and section 5 which was Employment, Wages and the Safety Net, and really there isn't an awful lot we want to point out from that, it is there to be examined in full, but what is interesting to us is our view that we can do better and in fact we have done better in the past.

PN1255

These are figures that have been calculated by a former member of the National Wage Case Benches, Professor Keith Hancock, who was former Senior Deputy President, and what is in my view again one of those things where if only, if only, and that is that growth rate in real average weekly earnings which occurred from the end of World War Two right through up until 1972, and we don't include the last two years which actually have an extraordinary growth rate, but we think that that was a growth rate well beyond the capabilities of this economy to sustain, and actually led to the subsequent period of 18 years of where there was virtually no increase in the real level of earnings across Australia.

PN1256

The important point here is that we have started finally once again to raise real earnings across the economy, and it is now growing by about 1.5 per cent, but I just point out that there have been periods in the past, and you can take the post-war periods, both World War One and World War Two, where you can see that from 1920 to '31 the average growth rate was 5.1 per cent, and the growth rate from 1947 to 1972 was 3.1 per cent, whereas the growth rate in 1992 to 2001 was a much more subdued, but still I think better than nothing, increase of 1.5 per cent.

PN1257

I now turn to make a few remarks about the drought. I suppose the single most important aspect of recognising that there has been a drought is to say that its effects do not disappear with the first rain. There has been certainly one disastrous crop season for a very large part of Australia, and whether this is the worst drought on record or the second worst or however you want to calculate it, this has been the drought we have had right now and it has had a devastating effect on a very large part of the Australian economy.

PN1258

The table that I note to the Commission, but I wouldn't ask you to turn to it, on the employment in the north and west of New South Wales versus the national employment rate, you have this huge growth in unemployment which jumps from 5.0 per cent in October 2002 to 9.1 per cent in January 2003 in the drought-affected areas of north and west New South Wales.

[11.24am]

PN1259

So here you have where the Australian unemployment rate has been more or less at a typical rate, in fact, it has not even actually fallen, yet you have in that segment of the economy - in the drought affected parts of the economy you have unemployment just virtually doubling, and what that itself says is there are repercussions that will affect other parts of those areas so that it is not just those farms, it is those farms and who those - output is bought from by those farms and it is the people who live in the surrounding towns and it is in the regions so that there is no localising the effects of this drought to the ends of the rain, nor to the end of the farming gate.

PN1260

They are going to be enduring an ongoing and that they will affect a very large number of people besides and beyond those who will be within the farming community itself.

PN1261

SENIOR DEPUTY PRESIDENT WATSON: Dr Kates, do you know whether any element of a growth in unemployment is a result of farmers seeking off farm employment?

PN1262

DR KATES: Well, I have no knowledge one way or the other, but I am certainly happy to look into that. Okay. But at the end of the day really what we are trying to point out and that is that the economy is to some extent moderating even now because of the drought, but that does not take away from the fact that the economy is moderating. This is something that I think is just recognised by all of us at the bar table and the ACTU has referred to it itself at the very beginning of its first written submission, so that you can say it is just the drought, but it is more than just the drought, but it is in fact - the drought has been extraordinarily important and it has affected the Australian economy, it has affected our rate of growth and it has affected our ability to export and it has affected our balance of payments and it is likely to have lasting repercussions on the Australian economy.

PN1263

I now turn to look at something we describe, perhaps grandiosely, as geo political risk, and this is found in section 7 of our written submission, first written submission, and no-one can go home and turn on their television set or turn on their radio without being reminded constantly, or pick up a newspaper without being reminded constantly that there are events going on that are having very destabilising effects on the economy, of the entire world of which Australia is going to share in that.

PN1264

I would simply draw attention to the statement made by the Reserve Bank governor, who is - whose last statement on monetary policy, published February, is reproduced in paragraph 7.23 and just a couple of things that he said that I think are important in understanding what is taking place kind of across the world, and he says to begin with:

PN1265

The weakening in world economic growth in the closing months of 2002, reinforced by the uncertainties of Iraq, was reflected in the three main developments in financial markets.

PN1266

Now the reason I think that is important is that he separates out the weakening in world economic growth from the actual war in Iraq. It is no the war in Iraq that has caused this. There is a weakening in world economic growth irrespective of the war in Iraq and that is important because the war in Iraq will itself have destabilising effects on our economy, on the economy as our trading partners and across the world so that we cannot think of isolating ourselves from those effects, and yet even if you exclude those effects he points out to the world - the weakening in world economic growth is taking place. He then points out that:

PN1267

It also lets the resumption of monetary easing by a number of central banks, including importantly both the US Federal Reserve and the European Central Bank. For the industrial countries as whole, official interest rates at an average levels not previously seen in the post war period.

PN1268

Because of the weakening of economies across the world they are lowering interest rates to levels unseen in the post war period. That is a circumstance untrue here in Australia. We have actually had a widening differential between interest rates here in Australia in comparison with interest rates across the world, so that the Reserve Bank's concern about inflation has led to a concern about maintaining interest rates higher, which has to some extent kept the Australia dollar higher than might otherwise have been which has to some extent added to the pressures on our current account, so that what we are looking at is a situation where the interest rates in Australia are already higher than elsewhere while they are being brought down, even now in other economies. Now I would also point out that he says:

PN1269

The main risk to a continuation of these favourable trends,

PN1270

which he notes exist in the economy,

PN1271

would arise if there were a significant deterioration in confidence about the future conditions associated with the international outcome.

PN1272

I think that the point we are trying to make is that this is a very likely thing to take place and as he said at the very end of this:

PN1273

At the time of the February board meeting it was clear that the international outlook had not improved, with incoming information generally confirming the loss of momentum in the major economies late last year and a general increase in geo political uncertainty. There is thus a risk that the recent international weakness will become more protracted and will weigh more significantly on growth in Australia than it has done to date.

PN1274

So the point we are trying to make in just noting these quotations from the Reserve Bank governor is that there are profound geo political risks that are facing us right now and they are well beyond the normal risks that one finds in an economy of this type.

PN1275

JUSTICE GIUDICE: Yes, we might adjourn for a few moments. Thanks, Dr Kates.

SHORT ADJOURNMENT [11.32am]

RESUMED [11.46am]

PN1276

JUSTICE GIUDICE: Thank you, Dr Kates.

PN1277

DR KATES: Thank you, your Honour. I just discussed the geo political risks and I would like to tender as an exhibit the front page story from this morning's Financial Review which has as its headline:

PN1278

Exports and housing to dent growth, building approvals fall four per cent, Iraq fuels uncertainty...

PN1279

which in my view provides a certain kind of up to the minute support for the submission we have been putting.

PN1280

JUSTICE GIUDICE: Should we give this any more weight than any other journalistic opinion, Dr Kates?

PN1281

DR KATES: It is always tricky, your Honour, but when they agree with us, I think it needs more weight.

EXHIBIT #ACCI5 ARTICLE FROM FINANCIAL REVIEW DATED 02/04/2003

PN1282

DR KATES: Anyway, as I say, it is from this morning's Financial Review and what you have here - this is a front page story and the first paragraph really goes right to the heart of what we have been trying to say and it reads:

PN1283

The slow-down in the Australian economy is occurring faster than expected with global weakness cutting into exports as the housing sector comes off the boil.

PN1284

And then it continues:

PN1285

The fragile international economy and the drought led to a bigger than anticipated $1.9 billion trade deficit in February, the 15th in a row. Exports led by coal and mineral fuels slumped six per cent and building approvals dropped four per cent for the month, led by a sharp fall in medium-density developments. Approvals have steadily declined since August last year and are now 30 per cent below their peak. The Reserve Bank of Australia board, which met yesterday, is widely expected to keep the official case rate unchanged at 4.7 per cent for the 10th consecutive month, but economists said domestic concerns, the war in Iraq and new weakness in the US economy might force the central bank to cut rates in the months ahead, despite signs that the drought is coming to an end. June bank-bill futures closed yesterday with an implied yield of 4.63 per cent, signally that investors believe there might be a cut by mid-year. The Australian dollar finished local trading and little changed at US60.3 cents. Federal Treasurer, Peter Costello, said exports of non-rural goods were pushed down by the weak world economy:

PN1286

He said:

PN1287

Imports of goods and services also fell in February, although imports of capital goods remained strong, reflecting the strength of domestic business investment spending, Mr Costello said. While the economy remained relatively healthy, both consumer and business confidence which point to future spending patterns have slumped because of the war and the dry weather. This, combined with the housing downturn, has prompted some economists to predict Australia's growth rate could fall below three per cent for 2003, down from the four per cent levels achieved in recent years. As a result, the number of market economists expecting a rate cut is now on par with the number who expect a rate rise.

PN1288

Quotation:

PN1289

The construction downturn means the Reserve Bank is more likely to lower interest rates than to raise them, J.P. Morgan, chief economist Andrew Pease said. Opposition Treasury spokesman Bob McMullen said the Government's mid-year forecast of four per cent economic growth in 2003 or 4 is looking increasingly shaky. Recent data from the United States, including confidence and employment figures show the world's biggest economy is still struggling, heightening expectations that there could be another interest rate cut there as well. Boston Federal Reserve Bank president Cathy Minehan said it appeared the US economy had softened and policy-makers should be especially vigilant, while the worst may be over, signs for improvement are scant, she said. Analysts in the US are worried the fragile recovery could be undermined if the war lasts longer than expected or if oil prices again shoot up. The world's largest economy is expected to grow by about 2.7 per cent in 2003, up from 2.4 per cent last year. Europe is even weaker, with economists expecting it to grow by just 1.2 per cent, while Japan will barely expand at all. The Bank of Japan's quarterly survey of business conditions and sentiment released yesterday showed that auto and machinery exporters which accounted for nearly 40 per cent of Japan's exports are increasingly gloomy about the global business environment. Analysts cautioned that sentiment might worsen, given that many firms responding to the survey had not fully factored in the impact of the war in Iraq. The six per cent fall in Australian exports largely reflect the drop in non-rural goods. Agricultural sales which have been hit by the dry weather were flat. The drought, one of the worse in the century, is expected to shave up to one percentage point off the growth rate this year due mainly to weaker export sales. However, rain in some parts of the country has raised hopes the dry spell is coming to an end, prompting predictions export growth could pick up by the end of the year. Imports during the month dropped two per cent, reflecting lower demand for goods such as clothing and food. On a trend basis, Australia's monthly trade deficit has been growing for the past two years through surging imports and a slump in exports. Commonwealth Bank chief economist Michael Blythe said the large trade deficits which detract from gross domestic product were being caused by a mix of drought, weak global demand, firm domestic demand and aircraft imports. Building approvals have also been trending lower since last year, as an end to the five-year housing boom approached.

PN1290

What we say about this is that this is just the most recent of a series of reports and newspaper articles and statements that highlight the fact that the economy this year is far from being the same kind of robust economy we were dealing with last year and if I could just - - - 

PN1291

JUSTICE GIUDICE: Isn't the suggestion though in that article that there may be reasons why the Reserve Bank would take some corrective action in the form of a reduction in interest rates?

PN1292

VICE PRESIDENT ROSS: Yes.

PN1293

JUSTICE GIUDICE: And that could be seen as effective to deal with some of those problems.

PN1294

DR KATES: My own view about interest rate adjustments is that it is a difference between pulling on a string and pushing on one, that the RBA can actually cause an economy to slow without hardly a ripple. If they want to slow an economy they can do it by raising rates. Lowering rates has not - does not have the equal and opposite effect and you can see that in the United States, which was noted here. They have been lowering rates in the US for the past year and they are at record low. The same thing in Europe, the same thing in Japan. They have brought rates down in Japan to virtually zero and yet they can't get the economy to grow.

PN1295

Our concern with the RBA for many years has been that the RBA will push rates up and slow growth, but reversing that doesn't do anything worth talking about to move things along again, and it takes a very long time for the growth they have cut off to sort of resume. So that what we are saying about the state of the economy is more or less that you cannot conclude, as was concluded in the decision last year, where on page 56 it said:

PN1296

There is no reason to conclude that world economic conditions pose a significant threat to the Australian economy in the next 12 months.

PN1297

I think the only possible conclusion one could have this year is that world economic conditions do pose a significant threat to the Australian economy. [11.55am]

PN1298

There is then finally, in the ACCI4 there is the last section there which, to my mind, is a very worrying development and that if you have a statement there which is an article taken down from the ABS and it just notes that the Treasurer has indicated that there may yet be a reason if the war continues longer than expected that the Federal budget may yet go into deficit, which would be I think an extremely dangerous move and I think the effect on the Australian economy would be one that would give us a great deal of concern and it would have, as it has had everywhere the deficits have occurred, would have an effect of slowing down economic growth, even beyond whatever is already built into the economy at the moment.

PN1299

If I could then move to deal with another aspect of our submission and that is to discuss our survey on the safety net. The ACTU has noted in its reply submission in paragraph 26 that:

PN1300

The only evidence which employer groups offer for the constant and repetitive assertions of negative employment impacts as a result of the ACTU's claim are the results of their members surveys.

PN1301

Now in our view we have provided quite a lot of other material to show that the effect of increasing wages and in particular the effect of the safety net, even last year, had been to slow employment growth and to have some negative effects on economic activity in general, but one of the points we tried to make in our reply written submission at pages 3.2 and 3.3 is to note that there is an immense difficulty to isolate cause and effect relationships within an economy.

PN1302

That the ACTU can simply say, "Well, where is the evidence", when in fact the evidence is often circumstantial, much of it is inferential, that it comes as a comparison with the kinds of growth rates one might otherwise have hoped to expect in different circumstances, so that it is difficult to find the DNA or the fingerprint or whatever it is that the ACTU seems to be looking for, but we note that there are in the economic literature three major methodologies that have been used to isolate the cause and effect relationships between certain events and certain subsequent events to make some kind of connection between the two as one to some extent attributed to the occurrence of the other.

PN1303

And as we note on those pages is that there is first - there is the regression analysis which is what the column - - - 

PN1304

JUSTICE GIUDICE: What pages are they, Dr Kates?

PN1305

DR KATES: Pardon me?

PN1306

JUSTICE GIUDICE: What pages are they?

PN1307

DR KATES: This is R3-2 and R3-3.

PN1308

JUSTICE GIUDICE: Thank you.

PN1309

DR KATES: And this is in sections - in "ACCI Survey on the Safety Net" and it is number 3 of our reply submission, section 3. And what we point out there is that there is the regression analysis which is the approach that has been used by the Commonwealth and in particular for the trim model. It is an approach we have not ourselves adopted at any stage, principally because we think that those kinds of models will not give you the localised cause and effect relationships that are desired.

PN1310

They actually, I don't think, work to actually give you the close connection that you need. That is not to say that there isn't a cause and effect relationship, I just think that it is difficult to establish them with massive models of the economy which are not designed, I don't think, to deal with these kinds of issues. The second form that is used is the econometric model and where you actually, rather than just dealing with the whole economy at once, you try to focus in on one aspect of the economy, and there is a phenomenal labyrinth of work that has been done in this area, of which the outstanding example is the Lewis and McDonald work here in Australia.

PN1311

There is work across the world on this because it is important to recognise and to estimate the connection between the cause and effect of increases in the cost of labour and the subsequent effect on employment of labour. But then there is a third approach and that is the one that we have also adopted and that is an economic survey. You try to find through asking business what they did, you try to find some kind of relationship between an event and businesses statements on how they reacted, and this occurs both inferentially looking at business survey data overall and in the direct questions that we ask, in this case, on the survey that we did on the survey on the safety net, both this year and last year.

PN1312

The ACTU has taken the trouble to provide our survey results to Professor Ian Gordon, who is a director of the Statistical Consulting Centre at the University of Melbourne, to review the results as we were presented in our first written submission, and in our submission what Professor Gordon has done is demonstrate that the figures are less precise than they would be if we had a larger response rate, which is a conclusion which we would agree with, but what he has not done, and this is the important issue to us, what he has not done is demonstrate that these figures do not provide useful and important information in determining the effect of granting increases through the safety net.

PN1313

And in fact in our view he has established the value of these figures and has helped set the contours of this right. Now in yesterday's transcript at paragraph 510 there was a supposed quotation that was taken from us, but it is in fact a statement from our written submission which is then followed by an interpolation by Mr Watson and then it is followed by the second statement and we have already asked that this be corrected, but it is important that it is understood what we did say, and what we said is that there is - because we have seen this ourselves in our own survey work, and we have seen it this year and last year and of course economic theory would support the likelihood that this has occurred, but that there is a flow on from the safety net decision and there are employment consequences.

PN1314

That is what we said, and then we added:

PN1315

The only question of interest is the accuracy with which the survey provides information about the importance of these consequences.

PN1316

And that is true, that if it is small and if it hardly creates any unemployment, then that is one thing. If it is large and has a very definite and important consequences, that is quite another. So the reason we do the - the survey says we do them is to establish how important, how significant is it when increases are granted through the safety net, and then - so the first point that Professor Gordon raises is that - he asks about the nature of our membership and he says, the weights we use, he says are unacceptable because they are national economic weights.

PN1317

ACCI is made up of every state based organisation in Australia. It is made up of virtually all of the major industry associations in Australia. When we do a survey it penetrates into the entire Australian economy. We get back results from every sector. We get back results from every size of firm and they come from every state, and when we conduct these surveys we actually have more knowledge about the structure of the economy than we do even about our own membership.

PN1318

[12.06pm]

PN1319

We couldn't even, I don't think, tell you what the membership structure of ACCI is, although we could use our sample from our surveys to get an estimate. But the important thing here is when we weigh these data, what we have ended up doing is taking a sample which is made up of responses of every sector of the Australian economy, and we have reconfigured it into the contours of the actual Australian economy.

PN1320

He says - as Professor Gordon says, the use of weights is common and a generally desirable step. And that is correct. If fact, it is not possible to do any survey of any genuine size without thinking in terms of building some kind of weighting mechanism into it. That virtually every ABS stats you look at will have a similar kind of weighting mechanism built into the underlying structure, where they will gather data from different parts of the economy, and then reassemble it in terms of the contours they know actually exist.

PN1321

The second point I would make is that when we send out these surveys, the point that Professor Gordon said is that people will pick and choose whether they will participate, depending on their interest in the subject. So that it might occur that people say, oh, look, I didn't do anything with the safety net, it didn't affect me so I won't bother answering. Or they will say, look, my answers will simply be of no use to actually establishing that this had a bad effect, so I won't answer this questionnaire.

PN1322

If you see how our survey is actually conducted, you will see that that is a very small, in fact we would say virtually non-existent concern. Because this is not a survey of the safety net per se: this is a survey that comes attached to a regular survey we do. It comes attached to our survey of investor confidence, and it is put at the very end of it, with no warning that it is coming. It does not show up in the covering sheet, it does not show up in the covering letter. So that people who are answering our survey are simply answering about a survey they think is about the state of the national economy. And then they get to the end, and there is another set of questions, asking about what happened in regard to the safety net last year.

PN1323

So that the bias, if there is a bias, would be what bias you think there might be in regard to people answering a questionnaire on the economy: what sort of characteristics they might have in regard to the safety net. I would say there simply is no connection at all, or none that I can actually think of, between people who would answer a survey on the economy, and then have particular characteristics in terms of how they would react to the safety net.

PN1324

SENIOR DEPUTY PRESIDENT WATSON: But Dr Kates, again there is a simple statistical method of assessing non-respondent bias, its existence or otherwise, and its effect on the findings. Isn't that the sort of action that should be taken, rather than speculation as to what the non-respondents might have thought? It was done by the British Low Pay Commission, and that found very different results within the non-respondent group as compared to those who had responded.

PN1325

DR KATES: That is right, and that is exactly the difference between our survey and theirs. And what their survey is - - - 

PN1326

SENIOR DEPUTY PRESIDENT WATSON: But ACCI didn't do the normal statistical technique of checking on non-respondent bias.

PN1327

DR KATES: No, my point is this. That the Low Pay Commission sends out a questionnaire, and it is a questionnaire that is specifically on what happened when minimum wages were adjusted in the UK: how did you react, what did your business do. So that businesses will look at this, and the point they reached was that businesses that had been affected more harshly were more likely to respond than businesses that did not, that were not affected as harshly.

PN1328

What we say is that our survey protects against that, because we are not asking them to answer a questionnaire on the safety net as such.

PN1329

SENIOR DEPUTY PRESIDENT WATSON: But that is entirely speculation, is it not, Dr Kates?

PN1330

DR KATES: Well, think about it.

PN1331

SENIOR DEPUTY PRESIDENT WATSON: And there is a method of addressing non-respondent bias.

PN1332

DR KATES: Well, I am just trying to explain why we think this is a problem that is not of concern to us. That those who are receiving and responding to that survey are not responding to a survey about the safety net. They are responding to a survey on the economy, as they have done every quarter for years on end. So they get the survey questionnaire; they fill it out; they send it back.

PN1333

SENIOR DEPUTY PRESIDENT WATSON: Can you be sure there is no firm that responded to this survey that has not previously responded to your economic survey because of the attachment at the end?

PN1334

DR KATES: It would require someone to have shown enough interest in the survey to have picked it up, and say, "I'm not going to do this." Go, flick, flick, flick, flick, and then go, "Ah, there is a bit here on the safety net, maybe I'll answer it after all." And what I think is that while you cannot rule this out as a possibility, it is very improbable. What is in fact the case is that what we have is a survey that is responded to in the typical way by the normal number of respondents. And what we find is that they have answered these questions as we have typically asked them, and then they have gone on, 95 per cent of them, and answered the last set of questions as well.

PN1335

So that we would say that the bias that exists, to the extent there is any bias at all, is the bias that would exist in the data between those who would answer a survey on the economy, and those who would not: rather than being, who would answer a survey on the safety net, rather than those who would not. This is just simply the normal response rate to a normal survey that we ask, and what we end up with is the usual outcome that we expect in terms of respondents, and respondent numbers, and the industries, and the cross-sectional data.

PN1336

So that I would think that in terms of this survey, there is no reason to be concerned about that bias. And I would add something else. That we don't do these surveys to deceive ourselves about the economy. We don't run a survey of investor confidence, and go to the trouble and the effort and expense of running the survey, just so we can throw sand in our own eyes.

PN1337

Our aim in doing these surveys is to get genuine and accurate data, as genuine and as accurate as we can get it, so that we know something of what is going on in the economy generally all the time.

PN1338

SENIOR DEPUTY PRESIDENT WATSON: Well isn't that all the more reason to apply the normal statistical processes?

PN1339

DR KATES: Well, we do, in our view, supply normal statistical processes. The only serious quibble that the ACTU has raised is they say the number of respondents could be higher, should be higher. And in terms of respondent bias, there is simply no reason, in our view, to believe that there is a bias between respondents to our survey, and some kind of correlation with being compelled to increase wages through the safety net. There is no correlation at all, as far as we can think.

[12.15pm]

PN1340

And for the second reason why we think that the data has validity, and that is because we then divide the data up between those who answered, yes, we flowed this increase on and those who have replied, no, we did not flow this increase on. When you look at the differences between the two sets of answers you can see that they are an entirely different world of respondents, that if there was this concern that Professor Gordon raised, that businesses were interested in helping their mates by giving egregious answers, saying that things were worse, even though it hadn't affected them, then you would see that.

PN1341

You would actually look at the data where they said, no, it is not - we have not paid any of our employees the safety net increase, but it had a terrible effect on employment in our industry. And you don't get that.

PN1342

SENIOR DEPUTY PRESIDENT WATSON: As occurred with the Motor Traders survey.

PN1343

DR KATES: No, actually I don't think it did occur in the Motor Traders survey. What you saw in the Motor Traders survey was that a safety net decision was granted in May, a survey was conducted in December and you found in a proportion of those businesses that by the time the questionnaire was asked they had no employees. Now, you cannot tell what happened between May and December to employment in that firm. It may well have been at the time that the questionnaire - or at the time the decision was handed down an increase took place, that there were employees in those firms.

PN1344

But by the time the questionnaire was distributed that there were not, or even you may have found that a business had been considering some kind of expansion even if it didn't have any employees at the time. But it had been considering some kind of expansion and when they came to the situation where looking at the growth and the cost of labour, which it had been an increase in the safety net of $18, which not only was $18 in that year, but sort of opened up new thoughts about potential growth in the safety net year on year, they may have thought differently about increasing their level of employment.

PN1345

So that even where at the time fo the survey someone had said that they had no employees, it is not proof against the fact that it did influence decisions and did have an affect on investment and employment in their firm. But we go on from there and we note that there are a number, in what our view is extremely positive statements - extremely positive statements about what our survey did. And, firstly, he says there is no suggestion of bias in the actual questions asked. He said, and I quote him:

PN1346

In general they are even handed and clear in their instructions.

PN1347

He had, in our view, some minor quibble about question 1, but I think that on the balance of considerations it was a very clear statement of what we wanted, but all it was asking was to separate out at question 1 whether they had passed on the increase or not, and if one is in a different camp because of that it doesn't matter an awful lot because you get the aggregate results anyway and they are what we are most interested in. So there is no reason to think that if that one was poorly worded, which we don't think is actually the case, that it made any difference in terms of the aggregate outcome, since we are interested mostly in was there flow-on, was there an employment effect.

PN1348

I then turn to a point that Mr Watson made repeatedly yesterday, that he emphasised this, that where we are talking about the issue of whether the result should be similar from one year to the next, and he quoted us saying:

PN1349

In general we would agree with that assertion but not entirely.

PN1350

Now, I was sort of hoping he would go on and read the next question because rather than having an answer to what was, in our view, a very clear point, he was simply trying to pour sarcasm on what is in fact a very straightforward, and I think, extremely telling point about the accuracy of the survey, because what we said - and this is at paragraph 3.23 - what we said was:

PN1351

In general we would agree with that assertion but not entirely.

PN1352

Question 1 asked, for example - quote:

PN1353

Did any of your employees receive an increase in wages directly because of the safety net decision.

PN1354

Now, for the questions on what was the flow-on, what was the employment effect, certainly from year to year there may be vast differences between how businesses may have reacted to a decision. But in terms of this question - in terms of the question on:

PN1355

Did any of your employees receive an increase in wages directly because of the safety net decision?

PN1356

It would have in our view have been a telling point against the accuracy of the survey had these figures not been similar, because the underlying population amongst employers has not changed. There is no difference between those who have been paying the safety net last year and those who would be paying the year before. The structure of business and the word respondency and all the other characteristics of firms that would have meant that the safety net decision was passed on would have been the same - virtually the same in both years. So that when we say we - in general we agree with the assertion, but not entirely, what we are trying to point out is no more than that.

PN1357

If we had had vastly different numbers in question 1 then we would accept that the survey had not been tracking accurately and it would have filled us with genuine concern that the other figures were not valid, because once that isn't showing the same kind of result as the previous year then you would have serious misgivings about the others. But since they did show various similar results it gave us, I think, good reasons for supporting the accuracy overall, that this thing was measuring truly what was happening out there in the labour market.

PN1358

Now, he also notes the issue of response rate of 30 per cent which he said was borderline. This is discussed in paragraph 324 and he said, quote:

PN1359

In my opinion a response rate of lower than 30 per cent means that the results of the survey must be regarded as unreliable due to possible response bias. The survey is on the borderline of this threshold.

PN1360

Now, we would, of course, like it to be larger, but as I note, we are not trying to deceive ourselves about what is going on in business. If we could get it larger we would get it larger. But it is on the borderline. It is not hopelessly out - the Low Pay Commission at 18 per cent and it publishes those results and they think they get valid conclusions from it. [12.25pm]

PN1361

We think we get the same. We think that the issue is one of precision, but not of general direction, and ultimate accuracy. The point then is raised on paragraph 3.25 of that is the issue of confidence limits. And this is, in my view, simply the most important part of what Prof Gordon said. Because what he did not say is that there is therefore no number. you cannot use these figures at all. What he said was, that the figures are imprecise. That you have wide confidence limits. And in statistical terms, what he is saying is nothing other than, when we have said the figure is 7.1 per cent, well it may well be much lower or much larger. And the money quote, from his statement is this, and I quote:

PN1362

The 95 per cent confidence interval for this percentage, assuming simple random sampling, is 3.8 per cent to 12.3 per cent.

PN1363

Now what he is saying here, and this is in regard to our question on the effect on level of employment is this. Our figure in the survey shows that the figure is 7.1 per cent. What he says is, well, given the low response rate, you can't really say it is 7.1 per cent. You have to think of it in terms of a band. It is some kind of range that is accurate figure. So that the range he is talking about is 3.8 per cent to 12.3 per cent. So what he is saying, that it is wrong for ACCI to say it is exactly 7.1 per cent. What he is saying is, that you have a high probability, that is 95 per cent probability of finding that number between 3.8 per cent and 12.3.

PN1364

Now, maybe it is 12.3. That would be an extremely high number, and much worse than our figures show. But suppose it is 3.8. Well, what I would say is this. That if it is 3.8, that itself is extremely significant. Because what it is saying is that in virtually one firm in 25 in Australia, in 3.8 per cent of firms across Australia, that employee numbers were lower because of the decision granted in the safety net last year. Now that is a very large number of firms, and that would have affected a very large number of employees.

PN1365

SENIOR DEPUTY PRESIDENT WATSON: But it may be that the 2.5 was in excess - in reality it would have been in excess of 7.1, which was in reality 3.8. So there were more with higher than lower, for all we know.

PN1366

DR KATES: I am sorry, is this the point - I don't follow that?

PN1367

SENIOR DEPUTY PRESIDENT WATSON: If it were 3.8, if the number of employees higher response was in fact higher than the 2.5, the higher end of the range, that might well exceed the number of firms with lower employment responses.

PN1368

DR KATES: Well, this is the point, I think, raised by the ACTU yesterday that - sorry, Monday already - two days ago. The band width, it is possible, that you couldn't have actually had an odd situation where we have had an increase in labour costs, and there was a net effect so that we ended up with an increased employment. But what you would have to do, in my view, is torture the statistics beyond recognition, that you would have to accept that when it says 7.1, it really is 3.8. And when it says, 2.5, it really is 7.1 or something like that.

PN1369

SENIOR DEPUTY PRESIDENT WATSON: Well, that is a possibility, an outside one, but we don't know what - wherein the standard lies.

PN1370

DR KATES: World of infinite possibilities. But if I go to next one, which is the full-time employment, which is the one I think is really important, you find that not a single respondent said that they had increased employment, not one. So that - I think the numbers are 0.0, 6.6. And it is the fact that on our survey results, you have a - that no matter what your coffer limits are, if no-one says that they had increased employment, there is no band, there no width, there is nothing. It is a nobody increased employment.

PN1371

And at the same time you have 6.6 per cent saying they reduced full-time employment. That if it turns out to be 3 or 9, that what this is telling us is that full-time employment was down, because nobody was increasing full-time employment as a result of a safety net, and many, although without knowing - without being able to say precisely how many, but many had reduced employment, full-time employment.

PN1372

VICE PRESIDENT LAWLER: Dr Kates, can I just put this proposition to you? If what you say is correct about the survey in effect being a survey that reaches all corners of the Australian economy, and is reflective of what is occurring in the economy, doesn't the fact that unemployment is falling at the moment, according to the ABS figures and participation rates are rising, don't those two matters together cast profound doubt upon the validity or the reliability or the correctness of your survey results? Because if your survey results are correct, the ABS figures as to unemployment and participation rates, are inexplicable.

PN1373

DR KATES: I completely agree with that. In this case - - - 

PN1374

VICE PRESIDENT LAWLER: So it is either the ABS is getting it right, or your surveys are flawed - and your survey is flawed, or your survey is getting it right and the ABS is flawed.

PN1375

DR KATES: Well, I would say that the figures up until the September quarter in the ABS numbers were consistent with ours. That there would have been growth in the labour market, even with the increase in the safety net last year. And there would have been an improvement in full-time employment. But what we tried to do, and this what we specifically asked business to do in answering this questionnaire was to say, try to just isolate the effect of what you did as a result of last year's safety net decision. Just try to do that.

PN1376

So that this doesn't say, what was your employment in your firm? This says, what did you do as a result of the safety net? And it is entirely possible that the firm may have intended to employ 10, but only ended up employing eight. Or that they would have done what appears to have been the case in the survey. And that is, that they would have reduced the number of full-time employees, and added on a number of part-time employees. So that what you see in this data are figures that suggest that you would get slowish growth in full-time employment, which is what we had up to the new data coming in, and you would have more rapid growth in part-time employment as businesses try to get their total cost reduction more closely aligned with the revenue they are receiving from their sales.

PN1377

So that this isn't saying they would have cut employment in general. It is saying that they would have cut - they would have had less employment as a result of the safety net, than they otherwise would have. And that is a different kind of - it is a different kind of outcome. Now, finally do take up this issue of response bias. And as Prof Gordon notes, he says:

PN1378

Response bias occurs when the respondents to a survey have different opinions and characteristics to those who do not respond.

PN1379

Now in our view it is not just that we say it is that the willingness to fill in a survey in economy is unrelated to one's decisions following the safety net review last year. We say more than this. We say that the question should actually be turned around to the ACTU. We have done this survey into the effects of the safety net and we have done it twice. And as far as I am concerned, the actual effect cannot be denied. The actual effect cannot be denied. There is flow-on. I mean that is what our surveys show, that is what AIG survey shows, that is what the retail motor industry shows. There are job losses, because of the safety net.

[12.35pm]

PN1380

We have provided estimates in answer to six properly asked questions where responded would have assumed they were merely answering questions on the economy and not in regard to the safety net. Virtually all those respondent on the economy answered the question on the safety net. Thus whatever bias there is, is the bias of who would not answer a survey in regard to the economy, not who would not answer a survey in regard to the safety net. So that it asks through Professor Gordon how can anyone assume that the non-response to the survey have the same opinions and characteristics as those who have responded?

PN1381

But we say the question should be turned around to the ACTU. What does it know that is different from what our survey results show? What can it tell us about the non-responding population to our survey? It wouldn't have given them any reason at all to think that that survey wasn't tracking more or less accurately. It has nothing. It has nothing. It says nothing. It has no evidence at all other than to say that because the figures are imprecise, therefore they cannot be used at all.

PN1382

And yet we know from the Low Pay Commission material, we know from the Low Pay Commission that even increases that occur just in the minimum wage lead to job losses in affected industries. That is undeniable. That is what their own material shows us. Here we are dealing with something that actually increases not just the minimum wage, but increases award rates that affect 25 per cent of the private sector of working population.

PN1383

And we would say that if the ACTU has some reason, some evidence, something they can put on the table to tell us that the non-responding parts of the economy, the other 70 per cent that were not in our survey, so that we didn't have 100 per cent response rate, if it can tell us something about the characteristic of those businesses and what happened there, that would be interesting. They have nothing to tell us. They simply want you to deny any relevance to a survey that is so consistent with what we have found in AIGs and the retail motor industry surveys that it has no basis for actually coming to any other conclusion than recognising that there are job losses.

PN1384

We may not be precise, we may not be exact as we would like to be in being able to tell you exactly how many businesses cut employment as a result, but we can tell you with an absolute certainty that there were businesses out there that ended up with fewer employees because of the safety net decision, and we can tell you with equal certainty that there are some people who are unemployed today or under-employed today who would have more secure jobs and would be working or would be in full time employment had that decision not occurred. That is what we would tell you. That is what our survey results show.

PN1385

SENIOR DEPUTY PRESIDENT WATSON: So all the survey is telling us is that there is some flow-on and some adverse employment effect; is that what you are saying?

PN1386

MR KATES: It certainly saying that, and we also think that within limits it is also telling us what proportion of firms in the private sector are responding in that way.

PN1387

SENIOR DEPUTY PRESIDENT WATSON: But that tells us nothing about numbers of employees.

PN1388

MR KATES: That is true. You can - I mean, if it is 7 per cent of firms - what did we say, 400,000 firms, 7 per cent, what is that, 28,000? - and one job per those 28,000 firms is 28,000 jobs. So that is assuming one job per firm, so 30,000; assume one-and-a-half, 45,000. I mean, sure, it doesn't tell you what happened to specific numbers and if we wanted to ask a survey in the more comprehensive way and get into the actual numbers, maybe we could get that.

PN1389

We are trying to give the overall sense of what is taking place in Australian businesses in response to the safety net decisions, and we are trying to demonstrate that, well, many of the points raised by Professor Gordon are valid and we wish we had more precision. What he has not done - very importantly what he has not done is shown that the results are invalid, all he has shown is that they have less precision than would be ideal.

PN1390

As we have discussed in our submission that there are different kinds of standards of proof. We are not trying to show a link between some new drug that is being introduced and some side-effects where you would want extraordinarily high accuracy over a long period of time. Nor are we trying in general to get some kind of time series where the kind of accuracy you need on the time series bases is extraordinarily high relative to what you need in a, kind of, getting a ball park figure. So that the standard of proof that we think is relevant and acceptable is entirely different for getting a sense of what has been the consequences rather than the data themselves. Now, I - - - 

PN1391

SENIOR DEPUTY PRESIDENT WATSON: But it is magnitudes of effect, is it not? And if all you are saying is the surveys show there is some flow-on - there is some adverse employment effect, well, the Commission in the most recent decision accepted that there might be some negative employment effect in those sectors of the economy with a high proportion of award rate earners, and has for some time accepted that there was some flow-on, but limited. So, in fact, all you are seeking to convince the Commission of, hasn't that already been accepted by the Commission in the past?

PN1392

MR KATES: Well, then I would go on to the next stage and say that these give some idea of what those proportions are.

PN1393

SENIOR DEPUTY PRESIDENT WATSON: They give what, sorry?

PN1394

MR KATES: Pardon?

PN1395

SENIOR DEPUTY PRESIDENT WATSON: They give what?

PN1396

MR KATES: Well, they give - I think it is very important in the first instance anyway to actually get to the stage where the ACTU must accept that there are job losses. I mean, they have resisted that up until I think this case and I am getting - - - 

PN1397

SENIOR DEPUTY PRESIDENT WATSON: We are arbitrating, not conciliating, so it doesn't require agreement of ACTU.

PN1398

MR KATES: Okay. Well, I understand that, but the next stage is to get some kind of order of magnitude. The data we have in the survey we think are accurate. We think that even if you happen to think that as an overall estimate of what goes on in the economy generally, we think that when you isolate it just down to those firms in which people are paid according to the award, these are telling you not just that it is some, but that the numbers are significant. And that given the confidence limits that Professor Gordon has stated on those data on employment where it is between 3.8 and 12.3, it may well be 12.3, it may well be an extraordinarily high proportion of firms in Australia that are experiencing a reduction in employment because of the increases granted through the safety net, or it may be 3.8, but I don't say "only 3.8" because I think 3.8 per cent is a very large number.

PN1399

I think if our figures had just shown 3.8 per cent, I would have said that is a number that requires serious consideration about what is going on, because if you are finding that across 3.8 per cent of firms in Australia employee numbers are being reduced, then that tells you something that is important. It tells you that the safety net has to be calibrated so that whatever good that is intended through raising the real incomes of employees on the award is balanced against the effect on their jobs and their job security.

[12.44pm]

PN1400

So that when we deal with that particular figure we are trying to say that there is serious concerns, and need for concern, about the effects of the safety net and how it will affect businesses. I think our numbers are generally right. I have no reason to think that we are not, when we do the survey, tracking truly, but we do accept what Professor Gordon said, and that is that there are wider confidence limits that we would prefer if we had the perfect survey, but again we emphasise what Professor Gordon has said, is that the figures may be as low as 3.8 or as high as 12.3 on that particular example.

PN1401

But what we are at pains to point out and what he would agree with, is that the best estimate given the survey result, is the 7.1 per cent that we have published. That is the best estimate we have of the actual proportion of firms that have reduced employee numbers.

PN1402

JUSTICE GIUDICE: Is that a convenient time for you, Dr Kates?

PN1403

DR KATES: Certainly is.

PN1404

JUSTICE GIUDICE: We will adjourn until 2.15.

LUNCHEON ADJOURNMENT [12.45pm]

RESUMED [2.15pm]

PN1405

JUSTICE GIUDICE: Thank you, Dr Kates.

PN1406

DR KATES: Thank you, your Honour. We were dealing with our survey on the safety net before the lunch break and demonstrating that in spite of the criticism of Professor Gordon that it is a valid survey and it provides useful information. If I could just briefly go to the data from the survey and the results that it showed. These are found in our original written submission which is section 11 and I am dealing with the survey results which are found on pages 11-11 through until page 11-19.

PN1407

Really what I want to point out is simply that the two aspects that we want to deal with are shown in these data. Question 1 asked:

PN1408

Did you increase wages directly because of the safety net?

PN1409

We find that on the aggregate of our survey it was similar in both years and in this year 41.2 per cent. The second question asked whether those on over award payments received an increase because of the increases granted through the safety net to those who were directly in receipt of that increase. We find that in both 2001 and 2002 there is a similar increase and the increase was 27.7 per cent. We then asked about non-award employees and that is the third question, and that is 12.6 per cent. We asked about the effect on employment and what you see there is that there was an increase in employment in this year in 2.5 per cent of firms and there was lower in 7.1.

PN1410

But if you go on to question 5 you find that in regard to full-time employment there was virtually no increase last year and absolutely no increase this year. It was 0.0 per cent - said that there had been an increase in the number of full-time employees, but that 7.5 per cent of respondents have said that they were - that they had reduced employee numbers in their firms because of last year's decision, and that is contrast to 6.0 per cent in the previous year, although we would not make any observation at all on the fact that it is higher.

PN1411

And then the last question is a question we did not ask last year, but we thought it perhaps timely to get some sort of a sense of what has been taking place over the entire six year run up to that stage of the safety net, and we asked - we just said:

PN1412

Increases in the safety net having been granted in each year since 1997, have these increases had any effect on your decision to employ?

PN1413

And we find that for 0.5 per cent of firms that employment levels were higher because of the safety net, but in another 19.1 per cent the decision - the effect of the decision had been to lower the level of employment.

PN1414

JUSTICE GIUDICE: What question number is this, Dr Kates?

PN1415

DR KATES: Pardon me?

PN1416

JUSTICE GIUDICE: What question number is this?

PN1417

DR KATES: That is question 6 and this is found on page 11/13 of our original written submission.

PN1418

JUSTICE GIUDICE: Yes, I see.

PN1419

DR KATES: And, of course, it is an ongoing aspect of the safety net which has made that the concern that it is. We then go on in the second part, and this is on page 11-14, and we asked where firms had said, no, there had been no increases in wages directly because of a safety net, which is why it is 100 per cent there, and we find that the flow-on effects have been slower, although you would, of course, not expect it to be zero since an increase being granted in the safety net can have effects on firms, even when they don't have the increase granted directly.

PN1420

Same thing with non-award employees, we find 5.2 per cent there. Level of employment in their firms. Here, of course, there are no employees, but that as we noted earlier the effect might have been delayed so that:

PN1421

Did the increase in the safety net have any effect on the level of employment in your own firm?

PN1422

1 per cent said yes, 3.1 per cent said that it was lower. Question 5 - - - 

PN1423

SENIOR DEPUTY PRESIDENT WATSON: Why would an employer increase the rates when they are not required to, if it resulted in lower employment than they would otherwise provide?

PN1424

DR KATES: Well, you may have an enterprise which has a number of employees, the people across the road have had an increase, the people down the street have had an increase, so that they grant the increase because of the pressures of competition, so that they are compelled, if they want to keep their own employees, to match increases being granted through the safety net. But at the same time by raising the wages on average being paid to employees they end up employing fewer. It doesn't happen in an overwhelming number of firms but it does appear to happen in a proportion of firms.

PN1425

So that going on to question 3, which asks about the increase in payment to non-award employees, you find that 5.2 per cent this year, 8.6 per cent last year. We then asked about the level of employment in their firm and 1 per cent stated that it was higher as a result, while 3.1 per cent said it was lower. We asked about full-time employment and none at all had increased employee numbers, whereas 3.1 per cent had had a reduction. And then finally asking across the entire period of the safety net since 1997 we find that no business had increased employment because of a - well, 6.3 per cent stated that employee numbers were lower because of the increases granted through the safety net.

PN1426

Now, we come to then the third division of this and this is in regard to firms in which safety net increases were granted. And here you see an entirely different set of characteristics, and what we find firstly:

PN1427

Did the increase get flowed on to employees receiving over award payments?

PN1428

We find that 56.1 per cent said that they had and 43.9 per cent said no, and amongst those 43.9 per cent are firms undoubtedly where there are no employees who are not in receipt of the safety net increase. So that some firms that just wouldn't have applied, but our questionnaire didn't have that for that division.
[2.23pm]

PN1429

In non-award employees we find that non-award employees had their wages increased in 39.7 per cent of firms. We then asked what the level, the effect on employment, and the results show that in 2.9 per cent of firms employment levels were higher in 2002 because of it, but 15.9 per cent of firms just said that the number of employees was lower because of the safety net increase. Question 5 then asked about full-time employment, and we note again that you can reduce full-time employees so that you will have a reduction in what we asked in question 5, and that can be translated to an increase on what we have in question 4, so that one full-time employee can end up as two part-time employees, each having say 12 hours.

PN1430

And what we find here is in question 5, that the increased employment did not occur in any firm at all, whereas in 17.4 per cent of firms there was a reduction in employment because of the safety net decision last year. And then finally the last question deals with what has happened across all the period since 1997 and we find that in such firms 2.9 per cent said it was higher than it otherwise would have been and 30.4 per cent stated that it was lower than it otherwise would have been.

PN1431

So, what we state in presenting these data is that there are employment effects, there are flow-on effects; they are not trivial, they are not insignificant, that as Professor Gordon pointed out we have wide ..... for them and so that any number there may well be much lower than the figure we have calculated. It may well be much higher. The best estimate, and that is what these data, the best estimate we can give you is that these are what has taken place, and they seem, given what we know from our own discussions with employers, and also through the results of other surveys, the AIG survey and the retail motor industry survey, that they are entirely plausible, that these are the kinds of outcomes that occur because of increases granted through the safety net.

PN1432

I would then, if I may, turn to the UK Low Pay Commission and the survey that it ran, and I would just like to make a number of observations about what those results said. This is found in our reply submission, and it is in appendix 4 and it is the Low Pay Commission Survey of Employers, and it is ACC3I3 Tab D. And I don't wish to dwell on this, other than to point out that the things that are found in the UK survey are the same kinds of results that are found in our own survey.

PN1433

And what we do in Australia is of course increase award rates right up through the scale, whereas what they have been doing in the UK is they have just introduced a minimum wage and they have been progressively increasing towards what they think would be some kind of acceptable minimum relative to their medium wage rate, and even so, the effect on employment, even though the proportion of employees who receive the increase is much lower, nevertheless what you see in these results is that there is an employment effect, and it is one that cannot be avoided.

PN1434

I also would note that the Low Pay Commission was quite content to accept the 13 per cent response rate which they stated was reasonable for a postal survey. This is under paragraph 4 of appendix 4, and I note that - - -

PN1435

JUSTICE GIUDICE: Behind tab C in ACCI3. Is that right?

PN1436

DR KATES: Tab C, I am sorry, your Honour, it is behind Tab C.

PN1437

JUSTICE GIUDICE: Yes. Although it is appendix 4 to the Low Pay Commission report.

PN1438

DR KATES: Okay. It is appendix 4 to the Low Pay Commission Report. I am sorry, your Honour.

PN1439

JUSTICE GIUDICE: Yes. No, that is - - -

PN1440

DR KATES: Anyway, I am now on what is page 226 of that, and I am discussing paragraph 5, where they - or paragraph 4 where they talk about the 13 per cent response rate, and I just note down that column where they have their table A4.1 response to survey, and the response rate in the best of the area sectors is for child care, where it reaches 25 per cent. Our own survey is 30 per cent. So that what we would simply note is that it is certainly possible to get useful, valid and interesting conclusions even from surveys which have a low response rate and in which the bias is certainly established because of the nature of the way that survey is conducted, which is different from how our own is conducted.

PN1441

What is I think very important is the number of things that are mentioned there, and then I take the Commission to paragraph 12, which is on page 230 of that report, and it says there that table A4.7:

PN1442

..shows that three-quarters of respondents affected by the October 2001 increase in the minimum wage said they had increased pay rates for higher grade staff to restore differentials.

PN1443

Now, this is simply the same thing as we say, which is about flow-on, that here in the - well, there in the United Kingdom where they are simply moving up the very lowest paid workers they find that in three-quarters of firms higher grade staff would find - would have their wages increased at the same time. Now, that is shown on that table, A4.7, where they give you all the data across all of the firms and then they say that the average is 75 per cent.

PN1444

[2.31pm]

PN1445

Now, what we say is that we are not just raising the minimum wage. We are raising rates right across the wage structure. And what is the outcome of that is that firms, if they feel pressure in the UK to maintain those differentials, which they do, then here in Australia the intensity when you are raising all award rates to maintain differentials, would be phenomenally more important and would lead to a more certain increase in the rates being paid to non-award employees and over-award employees than even exists in the United Kingdom. And yet, as you see there, it has been a very significant matter where it has created very significant pressure on businesses.

PN1446

SENIOR DEPUTY PRESIDENT WATSON: Dr Kates, do you have any response with the caveats on the survey results expressed by the Commission in paragraphs 6 and 9, which the ACTU drew our attention to?

PN1447

DR KATES: Well, I would make the following observation, your Honour. That neither of these considerations applies to the ACCI survey, for the reasons we have raised. It says firstly that respondents are more likely biased towards those most affected by the minimum wage. And what we noted this morning is that the survey that we do is not a survey about the minimum wage or safety net. Ours is a survey that we put out quarter after quarter for years on end, about the state of the national economy, and we get back responses.

PN1448

SENIOR DEPUTY PRESIDENT WATSON: I don't think you need to repeat that. It was really in relation to the survey you are now addressing.

PN1449

DR KATES: Sorry?

PN1450

SENIOR DEPUTY PRESIDENT WATSON: It is really in relation to the survey you are now addressing.

PN1451

DR KATES: Oh, yes. Yes, of course.

PN1452

SENIOR DEPUTY PRESIDENT WATSON: I think you have gone over all that ground.

PN1453

DR KATES: Yes, well they take that as one of those things that they are willing to say, well, look, okay, it may not be absolutely applicable to all of businesses across the UK. But we are interested as well in noting what happened to those firms where the increases took place. And of course that is what we are interested in as well, which is why, in our own survey we have - we break the results down by those who have passed on the increase, and those who had not. So that what would be a comparable set of data to what you see here is the data we have in our own survey, where people have answered, yes, we have pass - we have paid the safety net to employees in our own firm.

PN1454

Paragraph 15, which is page 231 of appendix 4 of the Low Pay Commission survey, deals with the issue of staffing. And it notes that 46 per cent of firms reported making a wide range of adjustments to staffing in response to that increase. And then they go on to say:

PN1455

The most likely action across all sectors was to reduce overall staff levels reported by 31 per cent of firms affected by the operating ...

PN1456

That is around one in eight of all respondents. Now that is shown in the table there, which is found on the next page, which is page 232 and table A4.10. And it asks about a number of effects on staffing in those firms where there were a clustering of low paid, so that it is not of course across the entire UK labour force. It is just in those firms and industries which are most heavily affected. And what they found was this. They asked, what happened to overall staffing levels, is the first question they asked.

PN1457

You find that it is 31 per cent of those firms, staffing levels decreased. And this is from a high of 50 per cent in the retail sector, to a low of 20 per cent in social care. They then ask about hours worked. So the next question was basic hours, and what happened, up or down? and they say - find that in 27 per cent of firms there was a reduction in the number of hours worked by such employees. The next thing asked about, what happened to overtime hours? And they find again that overtime fell in 23 per cent of these firms. They asked about what happened to overtime rates? Again, you find this decreased in 14 per cent of firms. And finally, they asked, what happens to non-wage benefits? So that are these increases in the minimum wage being funded by reductions in others?

PN1458

And it finds it has virtually no effect there, but still it did lead to a decrease in all the other benefits received in 7 per cent of firms. So what we say here, and we agree with the Senior Deputy President that this is, in a sense, an intentionally biased survey. We want to know what is happening in those specific firms. Or the Low Pay Commission wanted to know what happened to those specific firms affected. But what you find in those firms is that the overall level of employment has been brought down. The data in table A4.11, which is found on page 233, that too deals with a number of possible effects. You find that profitability decreased in 84 per cent of firms. Prices went up in 62 per cent of firms.

PN1459

There was the introduction of new technological processes in 15 per cent of firms. That there was use of unskilled and unqualified labour - occurred in 11 per cent of firms. So that there was a switch between more skilled and less skilled employment. So that there were repercussions in these specific firms. Table A4.13 which is found on page 235 deals with whether the effect on different age categories. And the one I think has some significance is that there is a greater likelihood to employ - sorry, there is a lower likelihood of employed workers aged 22 or over, in 15 per cent of firms, which is against a net 8 per cent in firms where it was more likely.

PN1460

So that really the conclusion we would reach is that where the minimum wage has been raised, the effect has been to lower employment, to cause prices to go up, to reduce the hours worked, to bring in labour saving technologies, all the things that we should be very careful to avoid here in Australia. And yet this is in an environment where we are raising the wages of the lowest paid, which only amounts to 3 to 4 per cent of the entire work-force. And what we are doing in Australia is raising the wages of a quarter of the work-force. So that we say these figures are extremely pertinent to our considerations here in Australia, but they must be, in our view, scaled up to reflect the greater breadth of the safety net in how things occur in Australia.

[2.40pm]

PN1461

Now, if I could now turn to another section of our submission and that is to deal with the academic discussion of the minimum wage and I would first just make a passing observation on the Dowrick and Quiggin paper which is the one tendered by the ACTU and the point I would make is this. The ACTU has traded that paper as if it has cut the Gordian Knot, that it is the final definitive statement on what takes place in enterprise when there are increases in the minimum wage. It says that you can answer paper after paper by a paragraph in Dowrick and Quiggin.

PN1462

It seems to have the view that when they have spoken, everyone else must now recognise that they have settled disputes on these issues which have been going on for untold years and I would say no more than this, that is one turns to the appendix of the Dowrick and Quiggin paper - sorry, the bibliography of the Dowrick and Quiggin paper which is supposed to be a survey of literature, you will see an enormous number of papers that deal with these issues. There are an enormous number of papers and the Dowrick and Quiggin is a working paper up on a university site. It is not the definitive answer.

PN1463

It is not even a refereed paper. It has not gone through the normal processes. What its appendix may or may not be saying about the Lewis and McDonald paper that we have tendered, even the ACTU said that they found it difficult to follow it themselves. It is not really I think in our province to say, well, we are going to take that one on faith. The Lewis and McDonald paper I might note has been published. That one has not, so that one can put some weight on it to the extent that they have been willing to put their own names to it and say that that is their own considered judgment, but we have very serious doubts and particularly given that the overwhelming preponderance of academic literature says that they are not right would lead us to the conclusion that if we are going to think about what academics have written about movements in the minimum wage, that one would simply give their paper very little weight given what others have said on the same area.

PN1464

We have put in ourselves a number of other papers. The ACTUs main comment on the papers we have put in was simply to say that they have heard these before. That is the fact and in our view it is very important that the is reminded of what these papers say and that there isn't the kind of view that the ACTU puts forward, but, in fact, the overwhelming majority opinion amongst academics is that the results of academic research show that increases in the minimum wage, increases in wages generally, do cause harm.

PN1465

I won't actually ask the Commission to go to these, but I will just remind the Commission that the Lewis and McDonald paper which is found at ACCI2 at tag D notes that the total elasticity of demand for labour with respect to real wages is approximately minus 0.8 and what does that mean? It means that an increase in real wages of one per cent leads to a fall in employment of .8 per cent so that there is not only a negative relationship between increases in wages and decreases in employment, but it is a large one.

PN1466

It is not trivial. It is not unimportant. It is actually quite significant and this is a paper that has been based on Australian data, on the Australian experience. The second paper we refer to is by Lewis and it is at ACCI2 and tag C and it is the economics of the minimum wage and what Lewis says is the impact of the minimum wage on total employment and unemployment is small, but the impact on low skilled, low paid workers is high and the point he was trying to make in this paper is if you forget about all those parts of the economy, where increases in the safety net or in the minimum wage don't affect anyone, which is a quite large part of the economy, of course, but only look at what happens to employees who are actually low paid and will be in receipt of an increase of the safety net, what you find is that increasing the safety net will lead to a reduction in employment.

PN1467

This point is further discussed by the paper we have in ACCI2 at tag F which is the James Wootton and Dawkins paper which is titled Minimum Wages and the Fallacy of the Inflated Denominator and what they are at pains to point out is exactly the same thing and that is that it is not appropriate to look at the effects on the economy overall and the labour market overall when you are just increasing the wages of a small segment. Where you should look to see the effects is at that segment of the labour market that actually has their wages affected and what it says on page 68 of that paper is that:

PN1468

An estimated employment impact of a minimum wage increase as low as minus 0.1 could be associated with the demand elasticity for employees actually affected by increases in the minimum wage of minus 0.92.

PN1469

In other words, even if the effect on the economy overall when you did this global macro-economic estimate was small and he isn't saying it is small, but even if it were, the effect on those particular employees might be extremely large and there are estimates that have been done, in this case United States has shown that this is the case. We have also brought forward two papers from the National Bureau of Economic Research in the United States which has been dealing with the minimum wages where that has become a very controversial issue and this first paper I refer to is at tag G and it is called the effects of minimum wages on the distribution of family incomes, a non-parametric analysis and if one just goes to the conclusion of this paper and to the very final paragraph on page 30, what it says is this:

PN1470

In our view, the answer we obtained to the question of whether minimum wage increases reduced the proportion of poor and low income families is a fairly resounding no. The evidence on both family income distributions and changes in incomes experienced by families indicates that minimum wages raised the incomes of some poor families, but that their net effect is to increase the proportion of families that are poor and near poor. Thus, it would appear that reductions in the proportions of families that are poor or near poor should not be counted among the potential benefits of minimum wages. Rather, minimum wages appear to increase the proportion of families that are poor or near poor, suggesting that the efficiency in equity effects the minimum wages point in the same negative direction.

PN1471

So that the kind of certainty that one might have that raising the minimum wage does some good is eroded and I think eroded to the point where the question really has to be asked whether it is not even an aggregate, netting out all the cross effects, is not even a negative on the incomes received by the low paid and their families in particular.

[2.50pm]

PN1472

If I could then turn to the issue of job security and this was a question that was raised at the beginning by your Honour, the President, and this is from section 12 of our original written submission and it is called job security versus increased wages and it basically is a discussion of a paper found at ACCI2, tag E and it is a paper by Jonathan Kelly, M.D.R. Evans and Peter Dawkins of the Melbourne Institute of Applied Economic and Social Research and it is entitled job security in the 1990s, how much is job security worth to employees and the important point to make about this paper in our view is that what it highlights and it highlights it very strongly is that for those on low incomes, they would, of course, love to have greater incomes, but that the concern for them is to actually maintain and continue the income they currently have so that if raising the minimum wage makes their jobs less secure, then they would rather forego those increases.

PN1473

Now, I suspect in many instances the employees involved are unaware that their jobs are being put in jeopardy by increases that occur through the safety net, but what is important here is to recognise that this data, which is quite cleverly in my view calculated, comes to a very important conclusion and that conclusion is found on page 5 of that paper by Kelly, Evans and Dawkins and what it says there is this. It says:

PN1474

Job security has a high value to workers, especially low paid workers. For them, security is equivalent to a very substantial part of their pay. Other examples of combinations of pay and security that are equally attractive to workers are given in the rows of table 1 which is shown on that page. The relative impact of security of income satisfaction decreased as salaries increase. This may imply that over the long run, with sustained economic growth and increase in their earnings, job security may become a less pressing issue.

PN1475

But what it says is that for those on low pay which is precisely the area where we are concentrating on in this safety net proceedings, for those employees the effect is actually a matter of very large concern and what we have is on page 12.6 of our first written submission. We have reproduced in a sense the table that is found in the exhibit and we have done it so that the point comes across I think more easily so that you can actually see the point being made by these authors and for those on $10 an hour in highly secure jobs, you would have to pay them $14.44 or give them an increase of 44.4 per cent for them to feel equally satisfied with accepting that higher pay in exchange for the lower security, so that the point this paper is making and I think it is a really important and extremely valid point is that we take away this choice for many employees by making their work less secure, by raising their wages and we are not raising their wages by anything like 44.4 per cent, not even in nominal terms, never mind in real terms and the effect is to jeopardise their jobs at the same time that they are getting this increase and yet as far as they personally are concerned, as this data shows, they would prefer to have more secure employment than actually receive the increases. Now I would like to turn to the issue of the impact of the claim.

PN1476

JUSTICE GIUDICE: Just before you do, Dr Kates, there is another part of that chapter that you haven't put in your oral submissions. It is the issue of the safety net adjustments being available without any enterprise based negotiation or any bargaining at the enterprise level. Is the assumption underlying that or the principle underlying that that there should be no increases for people who are unable to bargain?

PN1477

DR KATES: What we would say, your Honour, is that the statutory scheme that has been developed is intended to encourage individuals to move into bargaining areas, that a true safety net will be a safety net below which no-one can fall and we recognise that there should be a safety net at the bottom of the wage structure.

PN1478

JUSTICE GIUDICE: I think the Act goes further than that, doesn't it, in terms of a safety net? Doesn't it suggest the award system should be the safety net, not just the lowest point?

PN1479

DR KATES: Yes, it does, your Honour, and we don't disagree with that, that it is the award system that should be providing the safety net, but what we are concerned about is that we are looking at a process which, rather than seeing it as a safety net that underpins the entire structure of wages paid, we are concerned that it is now being turned into a system in which every year we will raise the wages so that those survivors of the labour force get if not a real increase, at least maintain the level of wage payments, maintain the real level of income and what concerns us about that is that if we are dealing with a safety net, we shouldn't be pushing the system and the wage increases through the safety net upwards every year in the approach that the ACTU brings to this Commission.

PN1480

JUSTICE GIUDICE: Looking at paragraph 12.4 of your submission which is on page 12-1, there seems to be two assumptions that might be implicit in that paragraph. One is that employees who are unable to bargain to increase productivity should be as it were treated differently for that reason. The other is that absent the bargaining process, productivity doesn't improve. I might be wrong about that.

[3.00pm]

PN1481

DR KATES: Well, I will deal with the second first. Productivity does improve in the economy and businesses have processes that allow productivity to rise even business by business. Even outside the bargaining processes there must be - every firm with award employees, most of them in any year - virtually all of them across this period of years will have done things to raise productivity. I don't think you can stand still.

PN1482

JUSTICE GIUDICE: I would have thought so, yes.

PN1483

DR KATES: Yes, so I don't think you can stand still in any competitive environment. So that businesses are always trying to raise productivity and whether you have a bargaining system or not it is taking place in firms. But what we do say is that in firms where bargaining happens you have a process in which the wages are adjusted and the wage growth is adjusted to some kind of recognition of what is going on in that business, and that the productivity that may occur will be as a phrase, lumpy, so that you may get a good productivity growth in some years and not so good in other years.

PN1484

JUSTICE GIUDICE: But this proposition seems to lead to the conclusion that employees would be better to resist productivity improvements and store them up to use them in exchange for bargained increases. If you have a free system in which productivity improvements are occurring without bargaining, why should that in some way affect the opportunity of the employees in that firm to get safety net increases, if that is all they get, or am I misunderstanding the submission? I may be.

PN1485

DR KATES: Are we talking about firms in which bargaining takes place?

PN1486

JUSTICE GIUDICE: No, we are talking about firms in which bargaining doesn't take place.

PN1487

DR KATES: Well, in such firms - - - 

PN1488

JUSTICE GIUDICE: Well, we are actually talking about employees who don't participate in bargaining and who are dependent on the safety net.

PN1489

DR KATES: Well, I would say that for employees who somehow resist the improvement in productivity their employers have a very short life span; that if you are in a competitive environment in which the business down the road or the business overseas has a different set of characteristics amongst its employees who are willing to pitch in and willing to try to craft a more productive business, then in an enterprise where this isn't happening they will not be able to maintain, I think, production levels. They will not be able to maintain sales and they will be ultimately forced out of business and I would think in relatively short order.

PN1490

You would need both businesses - the employees of both businesses to be complicit in some kind of abstract sense in keeping productivity low for both of them to behave in that way. I would say that overwhelmingly no business can survive if its employees cannot - do not do what they can to improve the bargaining - improve the productivity enterprise.

PN1491

JUSTICE GIUDICE: Yes, but it is a question of whether by - it is a question of how one encourages the spread of bargaining. The paragraph seems to me to convey the idea that there should be a discouragement of increases for employees who are unable to bargain or don't bargain, because the increases they get through the safety net are not accompanied by productivity increases.

PN1492

DR KATES: Don't know. I - - - 

PN1493

JUSTICE GIUDICE: That is what the second sentence says.

PN1494

DR KATES: Well, if I did say that I wanted to say that. What we would say is that in firms where there is bargaining and increases take place, that the give and take of the bargaining process allows certain changes that are productive to that enterprise to take place, that would be more difficult to achieve in firms where such bargaining does not take place. There are workplace changes, there are new skill developments, there are - the introduction of new technologies may require different roles for employees, lots of things that happen that the bargaining system allows to take place.

PN1495

It eases these changes at the workplace level. Whereas in an enterprise where you do not have a bargaining process as part of the enterprise wage setting system, that many of these changes can happen but you do not have a kind of bargaining process in which you can say I realise there is new training involved that you have to take, there is - there will be a new duty that you will have to be involved with, the certain kinds of things that happen as the normal give and take in a bargaining process.

PN1496

JUSTICE GIUDICE: But why couldn't those things be done, as I thought you had conceded you earlier, simply as a matter of course without bargaining? In other words there are no restriction on improvements and productivity.

PN1497

DR KATES: In some places I suspect they are and other places they wouldn't occur, and that the concern we have is that where they don't occur, or where there aren't even the productivity increases in any year to fund the increases in wages, what you end up with is the wages increased and an employer is left with the situation in which he must then consider where am I going to get the additional income to fund this increase if I haven't got the productivity this year, if I find the market is so difficult that I can't raise my prices, if I don't really have some investment that I am going to forego, there is just nothing there.

PN1498

One of the bases that an employer will turn to to actually fund the increase in wages to employees is by cutting back on employee numbers so that basic economic theory, the marginal product of the last employee is raised by the reduction in the number of employees in the enterprise. So that we, of course, would not argue that productivity growth doesn't occur in enterprises where there are only award employees, but what we do say is that the productivity growth would be inhibited to an important extent, and in some years where the productivity growth has not taken place, there is no mechanism through which the business can actually ameliorate the situation other than often times cutting back on employee numbers.

PN1499

SENIOR DEPUTY PRESIDENT WATSON: Well, where productivity is occurring, and you say it does, why shouldn't the low paid share in the improved living standards arising from productivity?

PN1500

DR KATES: We wouldn't argue with that at all. Where productivity has risen it is entirely our view that the low paid should share in that, but the problem with the safety net is that you can't make that discrimination between which are the firms where the increases can be afforded and which are the firms that you cannot.

PN1501

SENIOR DEPUTY PRESIDENT WATSON: So you give it to nobody?

PN1502

DR KATES: Pardon?

PN1503

SENIOR DEPUTY PRESIDENT WATSON: So you allow no low pay to share any benefit of productivity growth?

PN1504

DR KATES: Well, certainly what we are - what we have - what we are intent on encouraging the Commission to do is to recognise that there is a distribution of productivity improvements across industry from zero up to quite good, and that when increases take place through the safety net they will affect each of those firms, and while the average improvement might be all right, what you will find in terms of that distribution is a number of firms - a proportion of firms will find that increase unaffordable. And the larger the increase is in any year the more firms will find the increase unaffordable.

PN1505

So that you will find, even as productivity increases or other combinations are being made in one form after another, the effect on employment will be greater the larger the increase that is actually granted through the review. And our concern is to remind the Commission that the increase granted is the same for everybody, but the underlying productivity in any firm is extremely variable, and even if we go beyond that productivity and ask about the profitability in any firm, there are good years and bad years for many firms and a system that tries to maintain the real wagein all seasons, good and good, good and bad, will continuously take people out of the labour market because there are firms across the economy that simply cannot afford those increases and employees in those firms will either not be retained, they will have their hours cut or the employment increases that might have taken place, just won't happen.

[3.11pm]

PN1506

SENIOR DEPUTY PRESIDENT WATSON: One thing I can't work out, Dr Kates, is why your respondents to the survey, you say over seven years of living wage increases we have employed less people than we would have otherwise, have not, at some point, sat down with employees and said, well, look, we can increase our employment or avoid reductions if we achieve some productivity, can we do this, this and this. They seem to continue to persist on the basis of operating off the award rates of pay.

PN1507

DR KATES: Well, I think that the growth of the bargaining sector across Australia is evidence that, in many enterprises, this has taken place, that employers have sat down with their employees and said just that. Look, there is benefits for both of us, both for employees and for us and that we should move into a bargaining system and try to reach a kind of accommodation between us, in good years we share out and bad years we try to be a bit more restrained. But, essentially, there is some relationship between wages paid and the success of the business. Our hope is also that this will continue to expand, that there will be, as time goes by, more businesses moving through, recognising the advantages of the bargaining process and that will take place. But we are, in 2003, at a half way stage, probably two thirds of the way stage and we haven't got there yet.

PN1508

So what we are trying to say is that there are a lot of businesses that continue to employ and pay in relation to the award and there are a lot of employees who are paid according to the award and this is a process that you want to ensure will not cost people jobs, that it is done in a fashion that recognises that the greater the increase the more jobs are being put in jeopardy in those sectors where there is no ability to bargain. I would now, if I could, turn to the issue of the impact of the claim and this is found in our reply submission on pages 6.1 through and to 6.19, 6.19. And this is on the distributional considerations in award wage increases. We just - I will just go through this, relatively quickly and point out some of the problems we see in how this is approached by the ACTU, how we approach it, ourselves.

PN1509

We will withdraw the pages that the ACTU was so aggrieved about because we had not understood where they got the data from, but aside from that one minor issue, we will deal with the issue of the costing. Table 1, on 6-2, shows the ACTU methodology of calculating the increase across the economy. And basically it shows a 17.56 increase, $17.56 increase in the weighted cost increase in any firm. So you have 31.3 per cent who are permanent employees, 6.1 per cent non permanent full time, then the part time is divided again and then there is a junior and so when you weight these and go through it, the weighted increase of a $24.60 increase across the entire economy, in this sort of representative firm, is an increase of $17.56.

PN1510

They then take this, as table 2 shows, and that is on our page 6-5 of our reply submission, and they say, 21 per cent of the workforce receives the $17.56 increase and that is an average dollar increase for each worker of $3.69. They then relate that to average weekly total earnings of $696.50 and they get a figure of 0.53 per cent which is the addition caused by the - - - 

PN1511

MR WATSON: I am very loath to do this but, again, notwithstanding all of the correspondence that has flown backwards and forwards, again, Dr Kates says "average weekly total earnings" - - - 

PN1512

DR KATES: I am sorry.

PN1513

MR WATSON: - - - and it is wrong. Average weekly ordinary time earnings and we have told him about six times, now, where it comes from.

PN1514

DR KATES: Well, I beg your pardon. It is average weekly ordinary time earnings, but in - since in this case, you weren't complaining about the accuracy of the number, it is the only thing that is of concern is the percentage at the end, is 0.53 per cent. We agreed that if you do it this way that is the number you get and we had no issue with that. Table 3 is our sort of characterisation of using their approach to the public sector, and they have their table of zeros, this is our version of the table of zeros and since the increase goes to nobody or virtually nobody, whether it permanent, non permanent, full time, part time, junior, you name it, the effect on all employees, aggregate or individual, is zero.

PN1515

And the importance that we attach to that is precisely that if we are going to think of the impact of the claim, it is worse than useless to include people who do not get an increase. And in this case, we are talking about the public sector. The ACTU insists upon including the public sector data in the total, so that it is in this way underestimating the growth in the cost impact of the claim and whatever you may say about whether we call it ordinary time or total earnings, the actual fact of the matter is that it is, we say, improper to include them in the costing if we are trying to think of what has actually been the effect on the typical business.

PN1516

Now we have thrown in, just for interest, on pages 6-8 and 6.9, two examples of firms in which there isn't the exact proportion of full time and part time, permanent and non permanent, junior as you find in the economy overall. And we do this to remind the Commission that whatever the average may be, there are particular firms who are paying it. There is no average firm, in this very important sense, that what table 4 shows is a firm where there is 100 per cent employees receive the $24.60 that is being claimed by the ACTU. Using exactly the same procedure that the ACTU uses, you get to the final column and it is still $24.60. That is the increase, it has not been eroded.

PN1517

If you look at table 5 and it is the costing the claim, where we have a different kind of firm, 65 per cent full time permanent, 15 per cent full time non permanent, 10 per cent part time permanent, 10 per cent non permanent part time, what you see is when you do the weighted average there, it is still on average $23.12.

PN1518

What the ACTU can never get away from is this one unimpeachable fact. And that is where an increase takes place, if their claim is granted, that individual will get an increase of $24.60. And if you are someone on the minimum wage, you will then have an increase in your wage of 5.7 per cent. And 5.7 per cent in some firms may be affordable, and I can tell you in a lot of other firms, it will not be affordable.

[3.20pm]

PN1519

And so that this is all right as a kind of, let us see what we can calculate as a universal average across the economy. But in terms of the impact on individual firms, it does not come to grips with the fact that particular business must pay particular employees particular increases, and that in some circumstances they will find it excessively high and unaffordable. We then - - - 

PN1520

SENIOR DEPUTY PRESIDENT WATSON: Dr Kates, isn't the issue of calculating an aggregate addition to labour costs in order to assess aggregate macro-economic impacts, bearing in mind that you will also consider micro-sectoral effects amongst - as well as the macro impacts in considering a decision? And the cost estimates are directed to providing a base for assessing the macro impacts of the cost estimates, either of the Commonwealth or the ACTU? That is their purpose. The focus on the micro individual firm, really isn't relevant for that purpose, is it?

PN1521

DR KATES: Oh, yes, I agree, it is irrelevant for that purpose. And - - - 

PN1522

SENIOR DEPUTY PRESIDENT WATSON: Well, that is the only purpose for which it is being advanced, either by the Commonwealth or the ACTU.

PN1523

DR KATES: Yes, but we simply use that procedure to remind the Commission that it is not an average that takes place. That I have always great difficulty in dealt with macro economic aggregates affecting other macro economic aggregates. Because that is not how economics works. An economy actually has individuals down at particular circumstances who do particular things. And at the end of the day - - - 

PN1524

SENIOR DEPUTY PRESIDENT WATSON: And that all yields a macro economic output?

PN1525

DR KATES: Well, we add them all up, and then we look at one and the other, and say, oh, this is what happened in aggregate. How many employees there were, what is the average wage, or whatever. But when we are actually trying to think of cause and effect, we are trying to look down at what happens at a specific workplace, and what we are trying to draw attention to is that it is all very well - - - 

PN1526

SENIOR DEPUTY PRESIDENT WATSON: You can't stop there, can you? You can't just focus on the individual workplace without then looking at the impacts elsewhere leading to the macro impact? For example, you may have a firm that is struggling, it is a private company. The owner has been pouring in more capital because it has been sustaining losses for years, into what is, in hindsight, a lost cause. At the time of another safety net increase, he finally says, well that is the last straw, I am not going to pour any more money in, my 20 employees are unemployed, or it is under redundancy and they are out. But the competitor up the road might pick up all the business and employ an additional person, so the net effect is to - with some productivity gain in the employing firm, rather than the 20 which you were suggesting should be the focus.

PN1527

DR KATES: Yes, I - - - 

PN1528

SENIOR DEPUTY PRESIDENT WATSON: And then, of course, the economy is more complicated beyond that. There are further relationships arise as well. You seem to be suggesting we just stop at the first form.

PN1529

DR KATES: No, I - if we had what someone might consider full employment, and if we were in a situation where we said the economy is just bursting at the seams, and we think that real - we can afford a real wage increase, and that over in some place there were firms that were at the margin, and this pushed them over the edge, whereas other firms were still expanding at such a great rate, that those employees were absorbed into it, to some extent - - - 

PN1530

SENIOR DEPUTY PRESIDENT WATSON: Because they have taken over clientele of the company that has gone under.

PN1531

DR KATES: That is right. If we felt that the economy were such that we were looking at a situation where there were just two vacancies for every unemployed person, then I would have a very different attitude to that than I do, given that we have under-employment of something like 1.3 million, and there are those who are unemployed who are desperate to find work. So that - - - 

PN1532

SENIOR DEPUTY PRESIDENT WATSON: So what is the implication of that in saying if there is any job loss in any firm, which would not increase minimum wages?

PN1533

DR KATES: No, it is not any. And in fact what I was saying - or we are saying, is that you do not have an economy in which it really is available to us to squander jobs. That there aren't enough jobs around as it is. That one of the things that empirically is well understood, if you are say, over the age of 45, and you are in a certain sector of the economy, manufacturing in particular, is one that comes to mind, but you are over a certain age. And you lose your job. The possibility that you will never work again is very large. Very large. It is a very real possibility and really a major matter of concern to all of us.

PN1534

So that we are not so flush with employment, employment opportunities existing in the economy, that we can start squeezing people out. And even saying that, okay, there is - somebody disappears over there, but somebody gets a job over somewhere else. That we would be in a situation where you might well be condemning someone, someone in their 40s or early 50s, to be unemployed for the rest of their life, while someone who is say, under 30 gets a job. And what is really - - - 

PN1535

SENIOR DEPUTY PRESIDENT WATSON: But what is the implication of it for our role in determining - - - 

PN1536

DR KATES: Okay, what I see is the implication - - - 

PN1537

SENIOR DEPUTY PRESIDENT WATSON: If we find such a person, do we say, no, there shall be no increase in minimum wages? Well, what are you asking to draw from that focus on the individual employee or the individual firm, rather than working through the broader economic effects.

PN1538

DR KATES: Well, what we are - well, first of all the broader economic effects, as far as we can tell, given the date on hours worked, the data form our survey, is that, the net effect is negative. That the net effect that on jobs, of increasing the safety net, is negative. What - we go beyond that and say, that we are an economy in which there is still a very high level of unemployment. And when you start looking at all the other bits and pieces, with the under-employment and discouraged workers, and that, there is a vast reservoir of untapped potential, of individuals who want to do more.

PN1539

And what we are saying to you, is that the decision in the safety net should recognise that these are circumstances in which granting an increase that is excessive, and we certainly say the ACTU claim is excessive, will lead to more people ending up unemployed than we should even begin to consider tolerable. That we should - last year we came to the Commission and we said, $10. Now, we thought that struck a fine balance between what could be afforded and even recognising that too would put pressure on some businesses. Nevertheless we thought that would have been acceptable in the circumstances, given the tremendous prospects we had, and given the state of the economy, and given the world situation geo political.

[3.30pm]

PN1540

This year we come and we say, well, look, $18 was the increase granted last year and as far as we calculated, you gave the $10 and virtually gave the second $10 as well all in one year. We are saying that things are a lot stickier this year. It is going to be a lot more difficult, that growth is likely to slow and the risks to jobs are infinitely greater this year than they were last year. Last year we were looking forward to a fall in the unemployment rate which has eventuated. This year there are real concerns that it may not fall and that they may actually start rising again, so what we are saying is that in granting an increase, if you choose to grant increase, you must in our view, in our submission recognise the more fragile nature of the economy this year and you must recognise in our submission that an increase granted into this economy must be tempered by the recognition that individuals will lose their jobs and when they lose their jobs, we will certainly tot it all up and find them represented in the unemployment statistics if we ever get those figures right again, but they will be there, but the actual circumstances are that individuals are going to lose jobs in an economy that is not as robust as it was last year and that we must cut our cloth to suit the times.

PN1541

SENIOR DEPUTY PRESIDENT WATSON: That is just a function of the economy operating. Firms come in and out of business, jobs do come and go. If you applied that philosophy, you might even ban productivity at least until the short-term - short-term effects. In the long term, it might be detrimental to employment.

PN1542

DR KATES: Well, I am not of a mind to do that.

PN1543

SENIOR DEPUTY PRESIDENT WATSON: No, I didn't think so.

PN1544

DR KATES: But, you see, it is like in any month when you get the labour force data and they will say, you know, 20,000 jobs were created this month. Well, that is, of course, not true. Like, it will be 40,000 jobs are created and 20,000 jobs disappear so that you get a net. What we are concerned about, looking forward into the year ahead, is we are going to start looking at a very different situation where the net, jobs created, jobs lost, will be very close to zero and that if circumstances go very badly for us, although I don't forecast that at all, that you might actually start to see the net being more jobs lost in any month than are created. What we are endeavouring - - - 

PN1545

SENIOR DEPUTY PRESIDENT WATSON: Yes, I understand all that. It is really the focus on the individual firm of employees that was the issue that I was trying to get to.

PN1546

DR KATES: We then use more or less the ACTUs costing procedure to do our own estimate and we have taken the ACTUs figure for private sector costing and we have plugged it in ourselves and we see that in table 8 and again we use average weekly ordinary time earnings in table 9 and if you would turn to table 10 on page 6-15, you will see that this is our own ACCI private sector costings, so if we are going to look for a macro-economic number, this is the macro-economic number we would begin with. The first part, I just simply note that the first column shows 25.10 per cent which was the preliminary number produced by the ABS.

PN1547

The number is now 24.6 per cent. I also might add that in column 4, the figure is $17.44 and that is also the ACTU number for private sector costing which is different from their earlier figure which is for the total economy and that is why it is different, but essentially what you have here is you give the increase, the $17.44 increase which is the weighted total calculated by the ACTU. You give it to the core of the workforce which is 25.1 there or 24.6. The addition to earnings growth is $4.38 which is the quarter of 17.44. That is the proportion shown in column 1 of the figure shown in column 4 and that is shown in column 5 and that is $4.38.

PN1548

You relate that to private sector employed total earnings and it is $669 there and the total increase shown as a proportion is 0.65 per cent, so that if we are looking for what the effect on cost in the private sector are, we get a figure of around 0.65 per cent which I might note is the same as the kind of average growth rate that you see in the wage costs index in each September quarter, so over that flat, you get this increase of that sort of amount, but I wouldn't draw too much of a connection to that, other than to note that it is interesting they are the same.

PN1549

These figures cannot - whatever you want to say about flow-on and the rest, these figures are the rock bottom minimum. This is the figure of which no less than this amount will be the increase to private sector costs. We then add - in table 11 we just look at the flow-on effects and as Senior Deputy President Watson noted last year and we corrected for this, the figures have been halved from the data we find in our survey and what we find is that for putting - pushing the over award increases, the final column shows that that adds .18 per cent to the total costs, then you add non-award employees and that again from our survey is 6.3 per cent of employees which is a very small proportion of those employees and you find that it comes in the final column to 0.04 per cent.

PN1550

The total shown down the bottom of that final column in table 11 is 0.22 per cent, so that is in addition to private sector ordinary time earnings that occurs given our estimates of the flow-on and some crude, but I think relatively accurate figures for the movement. Table 12 shows our aggregate ACCI costs to the economy of granting the ACTU claim and what you see there is we have taken the 0.65 per cent which is the rock bottom minimum. We have added the over award, non-award increases and that the total increase of granting the ACTU claim would be 0.87 per cent.

PN1551

So this is not an invisible increase, it would not be minor and we say that it would be a very important addition to total cost. If I could then turn to deal with the issue of the minimum wage and just briefly like to talk about the NATSEM data which we, apparently, put some cold water on a couple of years ago. I am not entirely sure that I shouldn't have kept the view we had two years ago, but we are now under new management to this submission, so we have started to use it.

[3.38pm]

PN1552

The data that we really want to discuss are the national data found in our reply submission in section 5, and this is - and the tables are found at pages 5/3 and 5.5 in that reply submission. And really it is no more than to point out that there are effects on - not only on the employees but also on others. And the first of those tables - this is the one found in 5/3 - shows the estimate of poverty rates by labor force status for persons aged 15 and over. It is a very small point but I think it is an extremely relevant point, that your odds of being in poverty if you are unemployed are 50-50. If you lose your job your odds of being in poverty are 50-50. 57.5 percent it says in the table - it is even. You will be in poverty in half those circumstances.

PN1553

If you are not in the labor force then it is 16.7 per cent and that says you have one chance in six of being in poverty. If you are working part time it says 11.7 per cent. That is you have a one chance in ten of living in poverty. And then working full time the figure is 4.6, so that it is basically a long shot, you will be living in poverty one chance in twenty. And even then when you think of that one chance in twenty there is - for many, if not most people who are at that level of wages at some time in their life, they are not there permanently stuck at that point.

PN1554

The important issue is to maintain employment, keep people - to prevent people from losing their jobs, because just to use this kind of probability table, you go from having a one chance in twenty to be living in poverty to a one chance in two if you become unemployed, and that is a very large change in one's likelihood of being in poverty and one we say that in any decision the Commission should restrict - should ensure that that prospect is kept to the minimum. The next table is the one on estimates of poverty among individuals by main income source of family.

PN1555

The point that we make here is that if you run a business it is no guarantee of - to fortune, that in absolute numbers it may well be the case that, as it shows, 15 per cent of wage and salaries are living in poverty according to this estimate. But it says it is only 3.2 per cent of employees. But in contrast that - well, owning a business it is only 8 per cent. One business in seven - the owner of one business in seven, 14.3 per cent and in fact it is a very constant proportion across the five years shown on this table.

PN1556

What you find is that running a business - the five periods for 10 years on this table, sorry, your Honour, sorry about that - that between 1990 and 2000 this constant proportion of around one in seven of those whose family incomes are dependent on running a business are individuals who are living in poverty. Their income level is below the poverty rate and that is, I think, a very important reminder that when we transfer income through the safety net we are transferring from people who are running the business.

PN1557

Their houses are mortgaged, they are in debt up to their ears, that these are people who are putting tremendous amounts of their own capital and savings at risk in running a business and that for one in seven of these they are themselves living in poverty, or whatever the measure is. And that is I think a very important consideration when - considering the transfer from one to the other. If I could now I will turn to the issue of minimum wages.

PN1558

Sorry, I would like to go onto something else. I would just simply note that we can accept a floor on wages and can agree that if it is probably set it will provide some assistance to those on comparatively lower incomes, but we note that the minimum wage must be properly set with regard to economic conditions, business productivity and the likely effects on future growth rates.

PN1559

And in closing our discussion on the economy, I would like to just simply reiterate three points that were made by Mr Anderson, at the beginning, and these are these points. Firstly, economic growth in Australia has been and is moderating. The data we have shown and the forecasts we have brought to this Commission and the published material that we have discussed, all show that one conclusion and that is that the economic growth in Australia is slowing relative to the circumstances last year. Secondly, the point we make is that the domestic economic prospects are not as robust as they were at the time of the 2002 case. When we were discussing the safety net, a year ago, we were looking forward to an economy that was genuinely robust.

[3.44pm]

PN1560

We are no longer looking forward to an economy anything like that and in fact, our expectations, our own and of others, are for a year of weaker economic growth and of greater uncertainty. And thirdly, in relation to that, we note that the international economy is profoundly uncertain and is much more so than at the time of the 2002 case. The international economic situation has been weakening and that is irrespective of the war in Iraq. If one adds in the war in Iraq then we would say the uncertainty rises to a very high level and that we say that the decision in this case must recognise that and the other points that we have made. If the Commission pleases.

PN1561

JUSTICE GIUDICE: Thank you, Dr Kates. I just ask you a question which I think I know the answer to. You have not made any submission, in your written submissions or otherwise, about the proposal from the ACCER for an inquiry into the minimum wage. I take it that means you don't have any view on that issue?

PN1562

DR KATES: Yes, your Honour. Mr Anderson made a number of brief submissions, yesterday, on that, about the ACCER submissions and what he said is that there is a Parliamentary Inquiry now - - - 

PN1563

JUSTICE GIUDICE: That is right, yes, I remember that. Yes. And that it would be inappropriate for the Commission to do anything about it at this stage. Is that the thrust of it?

PN1564

DR KATES: That is correct.

PN1565

JUSTICE GIUDICE: Yes, I am sorry, I do recall that now.

PN1566

DR KATES: Okay. And we have a number of questions. We will try to endeavour to answer them and provide those answers to the ACTU before they begin. If the Commission pleases.

PN1567

JUSTICE GIUDICE: Mr Moir.

PN1568

MR MOIR: Thank you, your Honour. If it please the Commission, may I tender the joint submissions of the AI Group and the Engineering Employers Association of South Australia - - - 

PN1569

JUSTICE GIUDICE: Yes.

PN1570

MR MOIR: - - - dated 25 February 2003 and 26 March 2003, respectively.

EXHIBIT #AIG1 MAIN SUBMISSION DATED FEBRUARY 2003

EXHIBIT #AIG2 REPLY SUBMISSION DATED MARCH 2003

PN1571

MR MOIR: Thank you, your Honour. If the Commission pleases. This year's safety net review represents an important challenge to the Commission in terms of how it responds to Australia's current economic environment. Whilst there are signs of growth continuing, both the domestic and international outlooks are replete with uncertainties. The economy is confronted by the new realities of a war in Iraq, the impact of one of the worst ever droughts on record, falling share prices, uncertainty surrounding oil prices and rising input costs for businesses. [3.49pm]

PN1572

Together, these factors will dampen growth in the year ahead and can we say with respect that in this context, it matters little whether you are looking through the lens of the calendar year ahead or the financial year ahead. The risks moving forward are still there. Moreover, the balance of those risks is shifting increasingly toward the down-side. Against this background, we put forward this basic proposition. The capacity of the economy to afford a substantial labour cost increase which is not directly offset by productivity or efficiency gains is markedly reduced.

PN1573

In other words, the current environment reinforces the necessity for wage caution and wage restraint in adjusting the level of the safety net this time around. Contrary to the ACTUs assertions on this point, the amount sought in the claim does represent a very substantial amount of money and we oppose the claim on a number of key grounds. Firstly, we say that the claim is fundamentally out of step with weaker economic growth levels facing industry in 2003. The claim sits very uncomfortably with the present realities affecting the Australian economy and its outlook, including the possible impact of the war and the delayed impact of the drought.

PN1574

Everyone is now cautious in the face of these realities. Proprietors are cautious. Investors are cautious. Central banks are cautious. Everyone is cautious, except it seems the applicant in these proceedings. Second, we say that the ACTU claim if granted would deliver very little benefit to many needy low income households once the effects of taxation and a withdrawal of income support arrangements are taken into account. Third, both employers and low income earners would lose if a large safety net increase were granted and finally, the claim would if granted have a negative impact upon enterprise bargaining.

PN1575

The evidence we say demonstrates that there is a strong likelihood for a large safety net increase like that sought in these proceedings to fuel unrealistic wage claims and outcomes at the enterprise level. This poses additional risks in the light of the significant enterprise bargaining negotiations currently on foot in the manufacturing sector. Now, you would think that given the strong emphasis upon enterprise bargaining under the current statutory framework that the applicant might have something to say about this important issue.

PN1576

Regrettably, it barely rates a mention in the applicant's submissions and the mention it does receive is entirely unconvincing. The ACTU attempts to deny the undeniable, that safety net increases have a bearing upon wage negotiations at the enterprise level and I will return to this issue later on, but those then are the main grounds upon which we oppose the ACTU claim. Can we say also at this juncture that the position advanced by the State and Territory Governments for an $18 per week increase suffers from the same vices.

PN1577

Another increase of $18 on the back of last year's sizeable increase would represent a very substantial increase which the economy can ill afford at the present time. By contrast, our proposal for a moderate increase of $11 per week is economically sustainable. It will assist those low paid workers with unmet needs. It will deliver great bang for the employer's buck without lowering employment opportunities and it is consistent with the legislative framework, including the requirement to facilitate fair and effective bargaining and the desirability of attaining high levels of employment.

PN1578

A submission in this year's case also raises issues about the implementation of safety net increases and the need for the Commission to afford or should I say to conduct its own survey to determine the impact of such increases and I will go briefly to these points later on, but can I first turn to the economic conditions and prospects in a little more detail and why we say the ACTU claim is fundamentally out of step with the outlook for the Australian economy?

[3.55pm]

PN1579

To begin with, can I note that the ACTU does not seek to challenge our findings, that manufacturers have channelled back into the economy almost all of the returns from growth over previous quarters, by way of new investment, additional jobs and company and other taxes. The figures involved are quite telling. To the year ending September 2002, $10.3 billion in new investment had been generated by manufacturers. Thirty thousand new jobs, mostly full-time, and $3 billion in taxes. And this out of profits totalling 13.9 billion.

PN1580

Moreover, this process has delivered substantial benefits to workers and the community as a whole. It has helped us to this point, to sustain growth in the economy, thereby providing job security, job growth and higher living standards. In our respectful submission, it would be a tragedy to threaten this process for the sake of a few extra dollars, especially when the prospects for growth are now increasingly fragile. Excessive wage adjustments can destroy these benefits to workers, and the rest of the community, in future years.

PN1581

Turning to the prospects for growth, which we say are now increasingly fragile, economic forecasters are no longer debating whether the Australian economy will slow this year. Rather the issue has become. how much will growth be reduced by? This, of course, reflects the combined effects of a softer housing sector, the drought, continued global weakness, falling export levels, and uncertainty surrounding oil prices, and the impact of the war in Iraq.

PN1582

It is apparent from the latest national accounts data, and other sources, that the housing market has reached its peak, and that the beneficial effect of a low interest rate environment has now run its course in that sector. My learned friend, Mr Martin, on behalf of the State and Territory Governments conceded as much yesterday, when he stated at paragraph 727 of transcript, that his clients accepted there is going to be contraction in the housing sector.

PN1583

This leaves exports and business investment to hopefully fill the gap and help drive growth in the economy. However, business investment is likely to be influenced by the uncertainty surrounding the global and domestic outlooks. In addition, export levels have dramatically eased. In particular, manufacturing exports have declined quite sharply in recent quarters. And I refer in particular to the chart which is set out on page 16 of AIG1. Further, as that chart illustrates, there is little prospect of a return to growth in manufacturing exports until much later this year. And can I also say that the decline in export levels is further revealed by the latest trade deficit figures released yesterday, which show that exports fell by almost $600 million in the month of February 2003, with non-rural exports declining by 10 per cent.

PN1584

Now the explanation for this disturbing trend in export levels is of course quite straightforward. Our major trading partners are affected by weak global conditions, which in turn is reducing demand for Australian goods and services. The United States' economy is struggling to remain in positive growth territory. Europe remains stagnant, and Japan is, of course, in a chronic recession. And a significant appreciation in the value of the Australia dollar. In fact, since the beginning of this year, I understand that Australia has had one of the strongest currencies in the world, second only to the Syrian pound.

PN1585

So what do we have then? An uncertain environment, which is not favourable toward robust business investment, and exports that are dramatically weaker. And then, of course, there are the adverse conditions affecting the rural sector. Now, the negative impact of the drought upon economic growth in 2003 is fully canvassed in the submissions of other employer parties. However, it is important to note that the impact of the drought extends beyond just the rural sector. Our survey evidence shows that the drought is likely to lower production for one in three manufacturing firms, through weaker demand, resulting from reduced farm income.

[4.01pm]

PN1586

I refer in particular here to annexure A of AIG1. An additional one in eight firms, the majority being food manufacturers, expect input costs to rise as a result of the drought, so it is not accurate to say as the applicant appears to suggest that the impact of the drought is limited purely to a small segment of the economy. The drought will have significant downstream and upstream effects throughout the economy. Furthermore, as the Reserve Bank deputy governor has pointed out and this is set out in paragraph 3.17 of AIG1, the debilitating effects of the drought on farmers and their surrounding communities will only become more fully apparent in future quarters, that is when any safety net increase takes effect.

PN1587

Now, as if the drought wasn't bad enough, then there are the declines in the share market which according to the survey evidence set out in annexure A of AIG1 are likely to lead to lower investment, production and employment levels amongst manufacturing firms in 2003 and then finally to round it out, there is the uncertainty surrounding oil prices and the conflict in Iraq. Further survey work conducted by our organisation recently shows that even a short conflict in Iraq will act as a break on manufacturing growth and I refer in particular to paragraphs 2.12 and 2.14 of AIG2.

PN1588

A more prolonged conflict which now appears to be a quite possible scenario is anticipated to curb production and reduce employment and investment within manufacturing firms even further. The analysis conducted by the NAB which is set out in annexure D to AIG1 highlights the substantial economic risks associated with the war for the wider Australian economy. Now, in response, both the ACTU and the State and Territory Governments effectively argue that because no-one is in a position to identify the outcome of the war, then this is a neutral factor.

PN1589

The only thing we can say according to my learned friend, Mr Martin, yesterday is that the war is a source of uncertainty, yet that is exactly the reason why we say that caution is warranted in the current setting. There is little room for policy makers to get it wrong when you have, (a) a major military conflict, (b) a very severe drought, (c) a slowing housing market, (d) an appreciating Australian dollar threatening exports and then finally the reality of further and ongoing trade losses. The concern about the year ahead has we say been reflected in the caution adopted by the central banks around the world in setting interest rate policy. If I could take the bench to paragraph 3.32 of AIG1 which is the statement made by the Reserve Bank governor. If I could just quote that. Mr McFarlane had this to say:

PN1590

The caution shown by central banks such as ours derived in part from the downward revisions to world growth, but also in part from our suspicion that the risks were on the downside. In other words, if an outcome very different to the consensus were to occur, it would be more likely to be weaker than stronger.

PN1591

Now, that statement was made by the governor back on 6 December 2002 and if any circumstance has changed since then to alter this assessment about the need for caution, well, we would say clearly not and to the contrary, the balance of risks has continued to move towards the negative side since that statement was made back in early December.

PN1592

The ongoing need for caution is reflected in the more recent decision by the US Federal Reserve, just about a fortnight ago, to leave interest rates on hold. And in its press release, the Reserve pointed to the, quote, "unusually large uncertainties clouding the geo political situation". Despite the uncertainties and the reality of war in Iraq, the ACTU continues to found its case for a large safety net increase, on the premise that growth will strengthen or rebound in 2003/2004. However, as we note in our reply submission, how can they be so confident when central banks around the world are clearly not.

[4.06pm]

PN1593

Everyone is now cautious except the ACTU. The claim is also at odds with the sharp deterioration in labour productivity gains in recent quarters which, of course, make wage cost adjustments less affordable. And that sharp deterioration in labour productivity is depicted in chart 8 of our submission, that is at page 23 of AIG1. So to sum up, then, this part of our submission, we say that the ACTU claim is fundamentally out of step with weaker growth levels facing industry in the year ahead. Moreover, we say that the ACTU prediction of growth rebounding is highly questionable and should not be gambled upon by the Commission. The dicey nature of the ACTUs prediction, if I can use that term, is not alleviated by resort to a pettifogging distinction between this calendar year and the next financial year.

PN1594

The risks moving forward are still there, no matter what time frame one chooses. We note, also, that the ACTU has steered clear of our PMI results in this year's case. Now last year, the applicant made much about the PMI but not this time, it seems. And that is perhaps not surprising, given that the PMI results show a marked easing in growth over the past six months, reflecting higher input costs and a great cost price squeeze amongst manufacturing firms. Finally, in what can only be described as an act of near desperation, the applicant seeks to rely upon the recommendations made by the UK Low Pay Commission for a 7 per cent per annum increase over the next two years, in the minimum wage over there.

PN1595

We say that these recommendations have scant relevance to these proceedings. In the first place, the recommendations were made before the outbreak of the war and as far as we are aware, those recommendations are yet to be accepted by the UK Government. Also, the last time I checked, the UK was not affected by severe drought, maybe flooding from time to time but never drought. Moreover, though, is it any surprise that the UK Commission might recommend large increases in the minimum wage when the UK is just so far behind a country like ours, as the Commonwealth's submission helpfully reveals, at pages 36 to 38 or their original submission. The so called bite of the minimum wage in the UK is well below that of Australia and as my friend, Dr Kates pointed out, the award system in Australia also far more comprehensive in scope than the UKs minimum wage structure.

PN1596

SENIOR DEPUTY PRESIDENT WATSON: Mr Moir, the Low Pay Commission is proposing to increase the relationship to median earnings to around 50 per cent, is it not? Which brings it much closer to 54 on one of the measures cited in the Low Pay Commission report for Australia.

PN1597

MR MOIR: Two things I would say about that, your Honour. Firstly, even an adjustment to the level of 50 per cent of the median wage, it would still be substantially below the bite of the minimum wage in Australia, if you look at the data presented by the Commonwealth that I just referred to. But, moreover, I am responding to the proposition advanced by the applicant, that this Commission should take some comfort or reliance upon the recommendations which have been made in the past few weeks by the UK Low Pay Commission. We say that, on the issue of the quantum of any increase, there is simply no proper basis for a comparison between Australian and the UK or at least what this Bench should do and what the UK Low Pay Commission has just done. [4.11pm]

PN1598

Can I turn now to the second and third grounds upon which we oppose the ACTU claim; namely, that safety net increases deliver very little benefit to many needy low income households and that both employers and low income earners would lose if a large safety net increase were granted. The evidence presented in our submission establishes that on average low income households will receive around half of the extra costs imposed on employers as a result of any safety net increase, and this is due primarily to the interaction between the wages, taxation and family assistance systems. Indeed, many low income households will receive less than 30 per cent of the actual cost to the employer.

PN1599

This not only represents a very poor bang for the employers' buck, it also means that the persons for whom these proceedings are designed to assist, namely, low income earners, stand to gain very little direct benefit from a safety net increase, and certainly a large safety net increase, while simultaneously it reduces their employment opportunities due to the increased cost for employers. It is a Catch-22 situation of the kind referred to yesterday by Mr Martin. We say that the risks of such a lose lose outcome are augmented in the case of a large safety net increase.

PN1600

To illustrate the point we have shown in annexure J of AIG1 the degree to which the benefit of safety net wage increases are diluted relative to the cost borne by employers. Now, we have used the NATSEM data about EMTRs are effective marginal tax rates and that data records the degree to which private - or extra private income is eroded by the combined impacts of the income tests in the income support system and the income tax system.

PN1601

SENIOR DEPUTY PRESIDENT WATSON: Is that at a fixed point of time, Mr Moir?

PN1602

MR MOIR: I am sorry, your Honour?

PN1603

SENIOR DEPUTY PRESIDENT WATSON: Is that at a fixed point of time?

PN1604

MR MOIR: I would have to check that, your Honour.

PN1605

SENIOR DEPUTY PRESIDENT WATSON: Because the reality is you have various changes for the government arrangements annually and it would be more appropriate, would it not, to look at over the course of a year the effects rather than at one point of time where you might divorce wage movements from adjustments to government benefit arrangements.

PN1606

MR MOIR: Yes, I take your point and an issue has been raised about taking into account adjustments made in income support arrangements and the like, which I shall be coming to later on. But as for whether the NATSEM data arises from a fixed point in time I could check that and see what bearing it then might have upon what is said about it. But I was just going to point out that that data in its veracity has not been challenged. We have then taken that data and tried to measure the proportion of the full cost borne by the employer, including additional labour costs and compared that with what is actually received by the employee's household in the form of increased disposable income.

PN1607

We have termed that simply an actual benefit ratio, but you can call it whatever you like. Now, the charts in annexure J then show the EMTRs and the actual benefit ratios faced by the six different types of low income households, and importantly, these charts trace different levels of starting income. You are, therefore, able to see the actual benefit received for different levels of starting income within each household type, and that, we say, avoids the vice of using individual examples which the ACTU has been such a howling critic of in the past. What do these charts show? Well, they - - - 

PN1608

JUSTICE GIUDICE: Mr Moir, I think we might wait in suspension until tomorrow for the answer to that question.

PN1609

MR MOIR: Yes, yes, your Honour, I am happy to leave it on that question.

PN1610

VICE PRESIDENT ROSS: Mr Moir, just before we break can I draw to your attention a decision in print PR922761 which my associate will give you after we adjourn. It deals with this issue of retrospectivity and the matter that you raise in relation to the way the principles are interpreted. You refer to two first instance decisions and this is a Full Bench decision that deals with a related issue and I wanted to ask you some questions about it in due course.

PN1611

MR MOIR: Yes, your Honour. Is that the case cited by the ACTU - the re - - - 

PN1612

VICE PRESIDENT ROSS: I couldn't tell you whether it is or it isn't.

PN1613

MR MOIR: Re Victorian Shops Interim Award?

PN1614

VICE PRESIDENT ROSS: It is, yes.

PN1615

MR MOIR: Yes, I was going to address that.

PN1616

VICE PRESIDENT ROSS: Okay, that is fine, in your own time.

PN1617

MR MOIR: Yes, and, your Honour, you will find that - - - 

PN1618

VICE PRESIDENT ROSS: Well, in any event you can have a look at this one.

PN1619

MR MOIR: Yes.

PN1620

VICE PRESIDENT ROSS: And come to it in your own time.

PN1621

MR MOIR: Yes. You will find that issue - or that case is also addressed in AIG2.

PN1622

VICE PRESIDENT ROSS: Yes.

PN1623

JUSTICE GIUDICE: Could I indicate that it may be necessary to start a few minutes late tomorrow. It will only be five minutes or so but there is the possibility that might occur, but subject to that reservation we will adjourn until 10 o'clock tomorrow.

ADJOURNED UNTIL THURSDAY, 3 APRIL 2003 [4.17pm]

INDEX

LIST OF WITNESSES, EXHIBITS AND MFIs

EXHIBIT #ACCI5 ARTICLE FROM FINANCIAL REVIEW DATED 02/04/2003 PN1282
EXHIBIT #AIG1 MAIN SUBMISSION DATED FEBRUARY 2003 PN1571
EXHIBIT #AIG2 REPLY SUBMISSION DATED MARCH 2003 PN1571