Saturday, November 19, 2005

Why Treasury’s Got it Wrong on Tax Cuts

Back in the 1980s and 1990s New Zealand’s Treasury was notorious for functioning more like a rightwing pressure group than a government department: they were cheerleaders of the economic reform process; they published briefings with Orwellian titles like “Government Management”; and, eventually, many of their staff from that time did move on to join rightwing pressure groups or political parties. Theoretically, at least, this all came to an end some time in the late 1990s with Treasury – who, presumably, were somewhat chastened by the social and economic consequences of the reforms they championed – retreating from their role of small-government activists and starting to behave a bit more like a government department again.

That’s the theory at least; however, treasury’s most recent “Briefing to an Incoming Government” makes me wonder whether ideology hasn’t started to creep back into Number 1 the Terrace; either that or the people responsible for the briefing have a very strange grasp of public economics.

Particularly ideologically driven (or poorly thought out) is the recommendation in the briefing that the government reduce the top two personal tax rates. Doing this, according to the briefing, will have a “substantial [positive] growth impact”; a claim that is – supposedly – based on two things: economic theory and empirical evidence.

In terms of theory, the argument Treasury makes runs something like this: progressive taxation (increasing marginal tax rates), by reducing the take home proportion of higher income earners’ pay packets, provides a disincentive for working harder (or seeking promotion; or behaving in an entrepreneurial manner) which, in turn, leads to lower productivity and less growth. This theory is pretty much accepted unquestioningly by mainstream economists; yet as far as theories go it’s a pretty tenuous one; after all, it ignores the fact that, unless marginal tax rates are over 100%, workers still end up richer when they cross a tax threshold. It also ignores the fact that people may be motivated to work harder for more than just purely financial reasons. For example, they may be motivated to work harder by the desire for status that comes with promotion; or for the autonomy and decision making power that comes with being higher up the employment ladder; or by competition with co-workers (remember any promotion is still going to make them richer at the end of the day); or by professional integrity; or by the belief that their job actually helps attain greater social good (plenty of teachers feel this; yet apparently treasury officials are unaware of any such potential altruistic motivations. Why I wonder? Do they never feel them themselves?). On top of all this, as heterodox economists since John Kenneth Galbraith have noted, higher marginal tax rates may actually make people work harder, putting in the extra hours so they can purchase the goods they want. (For a good discussion of the motivations of economists see the last four paragraphs of this column; for a good discussion about theoretical arguments around marginal tax rates see here).

So the theory – then – is a little shaky; what about the evidence? According to the treasury report (on page 20):

[T]here is a…[rich] body of international studies that can inform and support our analysis. These studies have made significant advances in recent years, analysing the aggregate effect of taxation on the economy, and analysing the specific channels through which taxes impact on growth. Taken together, these studies strongly suggest that high marginal tax rates damage growth, though there is still some debate about the scale of this effect.
This I found interesting as my understanding about the relationship between taxation and growth was that there was very little evidence to suggest that countries with higher levels of taxation had lower levels of growth (economist Brian Easton makes this point here). Unfortunately, while the Treasury briefing refers to studies, it doesn’t provide any references for them (and Treasury haven’t – of yet – replied to my request for such references) so in trying to find where treasury got their data from I have been limited to using Proquest (an academic database). From Proquest I did find a few studies which claimed to have found associations between higher marginal tax rates and lower economic growth (hardly a rich body of evidence but, to be fair, Proquest may not contain everything that has been written on the topic).

Perhaps the most comprehensive of the studies I could find was Padovano and Galli in the journal, Economic Enquiry (Jan 2001). In this journal article Padovani and Galli claim to have found a correlation between lower marginal tax rates and higher economic growth. In doing so, they note that they are at odds with almost all other empirical studies on the topic of taxes and growth; however, they argue that most of these studies only use average tax levels and so do not capture the impact of higher marginal tax rates. Having now read through Padovani and Galli’s several times there are certainly aspects of it which seem a little questionable to me, including the fact that they estimate marginal tax rates for the countries involved (they do test this though), and also the way they do or don’t take into account the positive effects of government spending; but, to be honest, I don’t know nearly enough about econometrics to know if these questionable areas are significant or not. What I do know, however, is that cross country analyses are a fraught way of determining the effects of government policy (Dani Rodrik discusses this in this PDF file).

The alternative to cross country analysis is, of course, to examine the economic histories of particular countries; and once you do this it becomes pretty clear that, if there are growth benefits from reducing marginal tax rates, they are easily overshadowed by other economic factors. For example, in the United States the strongest period of growth in US history was in the 1960s, a time when the top marginal tax rate was 70 percent plus, likewise the soc-called long boom of the 1990s took place after Bill Clinton raised the top marginal tax rate (reference here) On the other hand, the post war decade with the lowest economic growth in the US was the 1980s, also the decade with the lowest top tax rates (for more discussions of the US context see here and here). Likewise, in Great Britain the top tax rate was cut from 60% - 40% with no discernable impact on economic growth (reference here). While in New Zealand the Labour Government’s 1984 tax cuts were followed by almost 10 years of economic decline and then a further 5 or so years of intermittent growth. And when a new Labour government raised taxes in 1999 the following 6 years where characterised by solid economic growth. Of course, none of this means that high marginal tax rates increase growth, or even that they don’t slow it down somewhat. But it does mean that any negative effect that they do have must be much smaller than other economic processes.

And, particularly in New Zealand’s case, other economic factors are important. If – for example – any tax cuts were to take place at present, with our economy almost at capacity, the consequence would, almost certainly, be a rise in interest rates. Something that would be best avoided if possible, given that New Zealand’s interest rates are already high by OECD standards and given that higher than average interest rates (everything else being equal) lead to currency appreciation, which harms exporters. (People who, if you haven’t noticed, our economy tends to rely on.)

In a similar vein, Brian Easton notes that poor fiscal policy, and its impact on currency rates, was probably the cause of New Zealand’s economic malaise in the 1980s and early 1990s. Which begs the question: do we really want to go through that again?

Of course, it would be possible a (as Treasury suggests latter in the briefing) to postpone tax cuts until the economy comes off the peak of its cycle, thus avoiding the need of interest rate rises. Yet at the same time, this also robs us of some of the potential for interest rate cuts (and, therefore, the chance of getting our interest rates back in line with the rest of the OECD).

All of which leads me to beleive that the economic benefits of Treasury’s tax cuts are far from clear. On the other hand, the costs are much more obvious. Despite talk of huge budget surpluses (which is just that – talk) New Zealand’s fiscal position leaves little room for significant tax cuts. Meaning that any such cuts have to come from expenditure. Something that Treasury fails to make clear in its briefing and something that I think “mainstream New Zealand” would be strongly opposed to. (My evidence for this belief is based on the number of political parties in the last election who campaigned loudly on cutting core services such as Health and Education. Answer: 0 parties which polled over 2%.) Of course, the natural conservative response to the question of where will the money come from? is that money won’t be cut from core services, but rather, “bureaucracy and hip-hop tours”. Unfortunately, the idea that cutting funding for the arts (Hip Hop tours etc.) could pay for substantial tax increases is absolute nonsense as I have discussed here. And, as for cutting bureaucracy, while small cuts could be made here and there (at a cost to services) there is very little significant that could go. When it comes to expenditure cuts Graeme Scott is bang on the money when he says that “it has to be done with a scalpel not an axe” (hat tip: Brian Easton) What’s more, given New Zealand’s aging population, and the associated health costs of this, in the long run, government spending is going to have to increase; something that tax cuts will leave little scope for.

None of this, however, features in Treasury’s analysis, nor do other impacts of the cuts like rising inequality. New Zealand is already one of the most unequal countries in the OECD and given that inequality has been shown to lead to increased violent crime and poor health outcomes as well as lower levels of happiness (a relevant link can be found here) it seems odd that treasury completely fails to discuss the distributional impacts of the proposed cuts.

All of which – like I noted at the beginning – appears to be evidence of one of two things: either Treasury is moving back to its 1980s role as an ideological pressure group, or they really don’t understand the fundamental public policy issues facing New Zealand.

Personally, I don’t know which is worse.

[Update: Sunday morning (yawn) after sleeping on this article I changed the first three words of paragraph 11 to "In a similar vien".]

[Update 2: Padovani became Padovano - thanks Genius]


Genius said...

the reference I believe is

Padovano, F. & Galli, E. (2001) "Tax Rates and Economic Growth in the OECD Countries", Economic Inquiry, 44 .

the different spelling of the first name might make it hard to find for others. (unless I'm mis-spelling it! Wouldn't want to put my food in my mouth)

Anyway amongst others

Mueller, D. C. & Stratmann, T. (2003), "The Economic Effects of Democratic Participation", Journal of Public Economics 87, 9-10, 2129-2155.

find a negitive tax vs GDP assocaition not that that counts for much by itself

anyway this is a review

Zagler, M. and Dürnecker, G. (2003), "Fiscal Policy and Economic Growth", Journal of Economic Surveys 17, 3, 397-418

amongst others...

Anyway the big overall is that generally one can say that certain tyes of expenditure and certain actual expenditures improve growth and others reduce growth (which is really what one would expect).

More expenditure on education for example will probably raise growth while more tax just spent against existing projects is very situation dependant.

Now I guess that to an extent agrees with what you wanted to say afterall treasury cant say "the evidence is we must cut taxes" when the evidence really says "depends on what expenditure you cut"....

Terence said...

Hi Genius,

thanks for the heads-up I'll change the spelling in a second.

I also found a few more articles today via Econlib: not that many though.

as for your other comments: I think I agree. Insomuch as that, while - holding everything else constant - a flatter tax system might lead to more growth, it is those things that get held constant that actually stop this from happening.