It something feels like you and I will never agree on anything, but I would have thought we could agree on what the numbers in the Grants Commission reports mean.

The Kennett government did not ''leave office with Victorian taxes still almost 7 per cent above the States' average''.

First, because the 7 per cent figure is the Commission's estimate of its revenue raising effort, not of the actual level of taxes. Actual tax levels in Victoria as reported in its last report were 4 per cent above the States' average.

Second, because those figures are for 1997-98. The Kennett government left office in 1999-2000.

Third, because since 1997-98 payroll tax has been cut by about $200 million a year and stamp duty by about $50 million, whereas in NSW and Queensland, as I understand it, taxes have risen in that time. I suspect that by the time the Commission calculates its numbers for 1999-2000, Victorian taxes will be shown as at or below the average for the states, and revenue raising effort just above it.

Then you complain that: ''Nobody has yet explained how taxes could be so high and at the same time spending on basic services could (according to The Age et al) have been cut to the bone. The answer is simple - such expenditure was not cut to the bone but was in fact running at around the average level for the states when Kennett left office.''

Sorry, Des, but the article I wrote on 6 November, and the accompanying tables, explained that clearly. I don't know why you are so resistant to learning anything from those you see as ideological enemies. I learn things from all sorts of people, even you! I was thinking of you in fact when I wrote that piece, and hoped that it would explain to you how the two went together. Apparently not, so I will try again.

1. Of course taxes can be high and spending low if a Government is running a surplus. The ABS data (5512.0) shows the Government in 1997-98 ran a current account surplus of $2.1 billion, and a general government surplus of $1.2 billion.

2. The Commission's data shows actual spending in 1997-98 was 10 per cent below the states' average, while state taxes were 4 per cent above average. It estimates standardised spending as 3 per cent below average, and revenue effort as 7 per cent above average.

3. Its measure of spending includes the $1.5 billion Victoria put aside for superannuation, much of which (dealing with the backlog) is essentially saving. On its figures, superannuation accounted for 10.6 per cent of ''spending'' in Victoria, cf 5.8 per cent in NSW and 4.5 per cent in Queensland. Take that out, and Victoria's actual spending was 14 per cent below average, and its standardised spending 7 per cent below average.

4. Social and community services spending was 10 per cent below average in actual terms, and 3 per cent below on a standardised basis. The reason it was relatively close to the average on a standardised basis is that it included the Government's spending on concessions, which was far above average, and essentially about delivering income support and subsidising prices rather than delivering services. Again, the Bureau figures show that concessions made up 9.5 per cent of social and community services spending in Victoria, cf 6.25 per cent in NSW and 5.5 per cent in Queensland.

I think my table summarised the Grants Commission data very well, frankly, and it should give the thinking economist the explanation you say is lacking. Of course it can be explained, because that paradox is precisely what the data shows! Surely we can agree on that.

Date: Thursday, 18 November 1999

Subject: IPE newsletter for November

As to the Grants Commission's methodology, having worked in Treasury for several years on Grants Commission assessments, written a few Treasury Submissions to the Commission, and been cross-examined at Commission hearings when leading the Treasury team I think I am reasonably well qualified to interpret Commission figures. I am afraid that I have to disagree with your interpretation of Victoria's relative tax and spending levels - and not for ideological reasons!

You are quite correct that actual collections of taxes per head in Victoria in 1997-98 were "only" 4 per cent above the average for the States. However, one point of policy focus is - by how much would average Victorian tax rates have to be reduced to bring them down to the average tax rates for the States? (One could argue that Victoria should aim to reduce tax rates to those in the State with the lowest rates ie aim for best practice).

The answer to that question is found by comparing actual Victorian tax revenue per head with its standardised tax revenue per capita, thereby obtaining the State's tax severity relative to the average. That comparison shows that Victorian tax rates were 6.8 per cent above the average tax rates for the States, equivalent in that year to about $550 million in "excess" taxation.

In fact, you acknowledge that Victoria's tax revenue raising "effort" (ie relative tax severity) was about 7 per cent above average in 1997-98. Subject to the next para, if that does not mean that Kennett left office "with Victorian taxes still almost 7 per cent above average", what does it mean?

Of course, Kennett left office in 1999-00. But for all practical comparative purposes 1998-99 was his last year, and figures for that year are not yet available. Looking at 1998-99, the reduction in Victorian taxes was only about $130 million, not the $250 million quoted by you (see Table 1.3 in Victorian Budget Statement No 2). Moreover, the NSW Budget for 1998-99 also decreased taxes by about $84 million, which would partly offset the relative comparison. It is unlikely that these tax changes themselves would have made a significant difference to Victoria's relative tax severity position in 1998-99.

On the spending side, again one should simply not compare actual levels of spending per head because there are differences between the States in the cost of providing services at any given level. Thus, to examine whether or not a State is spending above or below the average level for the States one has to compare its actual per head spending with its standardised per head spending.

In my Newsletter, I referred to spending on "basic" services not having been "cut to the bone". This was in part a reference to my article in The Age of 27 September (did you read it?) where I pointed out that, during its period in office, the Kennett Government did "no more than bring spending in key areas (schools, hospitals, social welfare and police) back to levels either just above the all-states average or only fractionally below". This statement was based on the following comparison of actual and standardised per capita spending levels in 1997-98:

Government Schools + 4.1

Police +2.6

Hospitals -1.4

Total Welfare +6.3

In short, in 1997-98 Victorian spending on the main services which were the subject of debate in and before the election was at a level similar to the average for the States. Again, it is unlikely that this relative position changed much in 1998-99.

It is true (as you point out) that total actual Victorian spending was 3 per cent below standardised ie below that needed to perform at the States' average. That implies that, overall, Victoria was "under-spending" (relative to the States average level) by about $500 million in 1997-98. However, that was not in the areas where the policy debate was focussed.

You suggest a couple of interpretations of Victorian spending which would imply less spending on services than a "straight" examination. However, these are outside the above key areas of basic services except for social and community services. Moreover, you don't mention the possibility that Victoria's lower spending may reflect a higher degree of efficiency in delivering services rather than a lower standard of services. That would certainly be true of hospitals where Victoria's case-mix system is almost certainly the most efficient.

In the case of social and community services, I am puzzled by your view that the provision of a "concession" is not a "service." Surely the whole object of expenditure in this area is to assist low income groups? The method of assistance will vary but a concession on (for example) electricity prices for low income groups is a "social service" just as is a disability pension.

I look forward to your article on Maddison's paper (I will make a point of reading it!). However, it was Ian Castles who was the critic of the handling of the UNDP paper by The Age and The Canberra Times. The Age journalist who wrote that up was Jason Koutsoukis. I was not criticising you for not attending the Symposium - I assume you are not the only one who could have covered it.

Date: Thursday, 23 November 1999

Subject: IPE newsletter for November

Further to our exchanges last week, I read your article on Maddison and note the welcome recognition it gives to the importance of institutional arrangements in explaining differences in economic performance.

You also quote Maddison as explaining Australia's slow-down after 1913 as reflecting the running out of easily mined gold, adding your own perspective that "the US industrialised". Later, however, you refer to the "American ascendancy" and quote Maddison as attributing that to (among other things) "its rapid development of market institutions".

Against this background (and notwithstanding your quote from Maddison about the running out of gold), one cannot help thinking that Australia's development of institutions after 1900 that were not market-oriented, particularly our labour market and protectionist institutions, played at least some role in our relative slippage in the growth stakes after 1913. Perhaps their remnants are even today contributing to our failure to "converge", at least until the last four or five years.

But, noting your reference to "home grown ideologues", I assume that you would consider any such suggestions to be "ideologically" based. (I'm sure that you didn't have me in mind in that reference!).

I note also that you did not take the opportunity to correct the UNDP report that an the increasing share of World GDP has been going to "rich" countries. As mentioned in my newsletter, that was uncritically reported in (among other papers) The Age. Ian Castles used the Maddison figures to point out the inaccuracies in that report.