Sunday, February 01, 2009

"Why targeted spending and concessions are preferable to outright tax cuts"

Fred Argy, at Troppo:

"Fiscal policy has become the subject of an intense ideological warfare among economists.

Over the long term - i.e. over the business cycle as a whole - economists do not agree on whether the structural budget should aim for a surplus or a deficit. This is understandable as several issues arise, such as whether governments can at times be more efficient than private enterprise, whether it is sound public finance for governments not to borrow, whether higher taxes are bad for economic growth, the equity premium puzzle etc.

But, until today, we thought we had wide agreement on whether governments should seek to smooth the effects of short-term fluctuations. It was generally accepted that we should be trying to iron out cyclical differences - through fiscal stabilizers (such as automatic changes in unemployment payments or in tax revenues) or through discretionary counter-cyclical fiscal measures (such as deliberate changes in tax rates or new expenditure initiatives). We have no such agreement..."

One group of economists – mostly libertarians like Robert Lucas, John Cochrane, Robert Barro and John Taylor – argue that the Government should go for tax cuts. Malcolm Turnbull – in many ways a social progressive on issues like abortion, gay rights, republic etc. - has now lined up with the economic libertarians, even though his views were mostly favorable to the $10.4 billion economic stimulus launched in October.

On the other hand, egalitarian economists like Paul Samuelson, Joseph Stiglitz, Robert Solow, Paul Krugman etc. want to use all relevant fiscal measures to minimize investment risk, ensure steadier income and avoid an increase in “core” unemployment rate – the rate consistent with inflation. This is also the view of some conservatives such as Martin Feldstein.

I find myself in the latter camp.

Our argument is that supply-side models are adequate models of the long run but they do not explain demand-side short-term economic fluctuations very well. Libertarian economic analysis is largely based on longer term models (e.g. 1955-2006 in the case of Makiw and earlier still by Barro), which allow for supply-side responses. This analysis does not apply to near-depressions.

On the other hand, models which include wage and price rigidities, such as New Keynesian models, do have a greater ability to explain short term fluctuations in a severe recession and can best address these macro-economic fluctuations.

This is particularly so when we are facing the prospect of a near-depression. In such a climate, monetary policy cannot work effectively where there is a break-down of trust among key players in the financial system and extreme risk aversion. It needs to be supplemented by fiscal policy.

The IMF is now urging its members to devise packages that “provide maximum fiscal boost to demand very soon”. If we relied on general tax cuts, it is clear that much of the tax savings would be saved in the short run. Public spending can provide much more bang for the buck than tax cuts. Moreover, tax benefits, especially if they are “permanent, pervasive and predictable”, cannot be later reversed once the depression is over. What we will then hear is how raising taxes will tend to kill growth and cost us jobs. Libertarians are justified to make an argument against public spending in general - but not at this point of the cycle.

What we need are selective, easily reversible spending cuts such as –

- bricks and mortar investments in “shovel ready” (no long lead times) infrastructure;

- time-specific jobless (unemployment) benefits;

- temporary tax benefits for low-income (genuinely low-income) persons to limit demands for wage increases;

- some assistance to people offering welfare benefits;

- through the states, assistance for all specific industries or regions that will bear the brunt of job losses;

- a boost in education and training benefits (health may need to proceed more slowly because it takes time to retrain medical trainees); and

- temporary tax benefits to encourage new investments and new technology (such as reducing energy or water consumption).

Such measures will only “crowd out” private investment and net imports if:

- there is pressure on productive resources;

- direct interest rate responses are likely to be very large, despite an accommodating monetary policy;

- exchange rate movements are correspondingly large; and

- there is significant forward looking consumption smoothing by private agents (Ricardian equivalence effects where people are smart enough to recognize that higher deficit spending will lead to higher taxes later).

The first only applies when the economy is operating close to full employment.

The second and third are crucial. The effectiveness of fiscal policy is only seriously damaged where (a) the smallest rise in investment and housing demand sharply pushes up rates of interest or (b) if the demand for real money is very insensitive to interest rate changes. But such extreme circumstances – and their effects on exchange rate movements, especially if other countries are doing the same thing - are unrealistic and unsupported by the evidence.

As for Ricardian effects, past experience suggests that any effects tend to be long term and then only dampen the initial impact by a third to one half.

We are also blessed with a stronger banking system and sounder government balance sheets than in other countries. This too is an important issue to consider.