Thursday, June 7, 2012

The Difficulty in Accepting "Discretionary Stabilizers"

As the debate on the appropriateness of doing more fiscal and monetary stimulus continues, here is a reflection on something that I never understood in that debate: even those who strongly oppose the idea of additional stimulus rarely question the workings of automatic stabilizers. This is surprising because under most economic models, a change in taxes or fiscal spending has a similar effect on GDP and unemployment whether this is built into the tax code or it is a discretionary decision of the government (this is even more true if everyone expects the government to react like that during recessions). Somehow, automatic stabilizers are seen as normal and effective (here is a 2008 WSJ piece by John Taylor arguing that automatic stabilizers should be allowed to work). But discretionary changes are ineffective and can only do harm.

The moment one admits that there is a good reason to have automatic stabilizers, then the discussion on how large they should be and how they should be complemented with discretionary fiscal policy is a natural one and cannot be avoided. Automatic stabilizers are not equal across countries and there is strong evidence that countries with smaller automatic stabilizers use discretionary fiscal policy more aggressively to compensate for this (which makes sense from an economic point of view). A few months ago we wrote a paper on this: the similarities between automatic stabilizers and discretionary fiscal policy. In addition to presenting evidence of how automatic stabilizers and discretionary policy can be seen as substitutes, we also show that among OECD economies the largest automatic stabilizer comes from stable government spending in the presence of falling GDP. In particular, in countries where government spending is large (and stable) we see strong automatic stabilizers. Finally, we also estimate fiscal policy multipliers but we do so by taking into account our argument that automatic stabilizers and discretionary policy are part of the same policy.

Antonio Fatás

Monday, June 4, 2012

Fiscal Federation or Fiscal Policy

Can a monetary union work without being a fiscal federation? This has been a question that has been repeatedly asked since EMU became a project. Economists consider a fiscal federation a way to share risks through a common (large) budget. Risk sharing allows for smoothing of business cycles at the regional or national level. In the absence of monetary policy (and exchange rates) the theory says that adjustment must come from either price adjustments (internal devaluation), labor mobility (towards regions/countries that are doing well) or fiscal transfers through a common budget.

Back in 1998 when EMU was about to be launched I wrote an article with the title "Does EMU need a fiscal federation?" My conclusion was that while it would be nice to have one, the costs of not having one where not that high (and the implementation costs seemed too high at that point). The main argument of that paper was that business cycles have become so synchronized that the costs of losing national monetary policies was small (there was limited risk to insure at the national level).

Has the current crisis changed the logic of that analysis? Clearly this crisis is more asymmetric than all the previous ones in the Euro area. The 92/93 recession was very similar across countries. And 2001-2003 was a period of low growth for some of these economies but there was no deep recession in any Euro country. But since 2008 we are witnessing a recession which is very large and is not spread evenly among all Euro countries. No doubt that the benefits of any mechanism to share risks and alleviate national business cycles looks more important today than in any of the previous crisis.

But how much risk is being shared through a fiscal federation/common budget? Paul Krugman makes a comparison between Florida and Spain (as two regions/countries that have suffered a real estate boom and are now looking for help) and argues that Florida has received in the years 2007-10 about $31 billion from "Washington" via the federal budget and programs. These funds are a "a transfer, not a loan". Spain has not received any significant transfer from other Euro countries because of the small and unresponsive EU budget. $31 billion is a large number but my reading is that this number overestimates the benefits of the US federal budget.

Here is my reading of the same numbers: most of the funds behind the $31 billion figure come from the lower tax payment that Florida did to the federal budget during 2007-10: about $25 billion. But there is no risk sharing (and certainly not a transfer) if all states decrease their contribution to the budget by the same amount. A quick look at the data says that Florida tax revenues went down by 12% while overall tax revenues went down by 8.4%. So there is some asymmetry, but this asymmetry does not amount to $25 billion. What is happening is that the federal government is running a large deficit. This deficit is the main way in which tax revenues are being smoothed in Florida. For this, you do not need a fiscal federation, you just need a government (national is good enough) that is allowed to run those deficits. The difference between 8.4% and 12% can be thought of as risk-sharing although this might not be a "permanent" transfer. Unless we believe that this is a permanent change in Florida's GDP, we expect this pattern to be reversed when Florida's growth is higher than the average (possibly it already happened in the period before 2007). So this is a transfer in bad times that originates in a payment that Florida did during good times to the Federal system (once again, assuming this is just a transitory event). In other words, this is insurance, which is great to have but it is smaller than the smoothing provided by the fact that the federal government is running a large deficit.

In conclusion, countercyclical fiscal policy is good whether it happens at the local or federal level and this is the main reason for the large swing in tax revenues in Florida. Yes, there is in addition some risk-sharing between states (and unemployment benefits are even a better example, but those should also be measured relative to other states and not in isolation) which only happens via the federal budget. This mechanism is indeed absent in Europe. So can European countries do in the absence of a fiscal federation? They should rely more on standard countercyclical policy (which they do via their stronger automatic stabilizers) and possibly use additional mechanisms that take the forms of loans (as opposed to transfers/insurance). Credit in bad times is another form of achieving some smoothing of business cycles. Unfortunately we are doing the opposite: we are seeing a combination of austerity-by-faith policy combined with governments being cut access from financial markets. Under these circumstances, fiscal policy has become procyclical, exactly the opposite than what many countries need. In the absence of proper countercyclical fiscal policy any other mechanism, including a fiscal federation, would have been of great help.

Antonio Fatás