The Stability and Growth Pact of 1997 permitted Eurozone countries to run a general government deficit of 3% of GDP. The only economic innovation of the Fiscal Treaty is the replacement of this deficit measure by one that restricts the “structural budget deficit”—the deficit that that would be incurred if the economy was operating at full capacity—to 0.5% of GDP where the national debt is greater than 60% of GDP, and 1% otherwise.
The measurement of the structural budget deficit is to be cyclically adjusted, which will permit larger deficits in times of economic recession. So far from outlawing Keynesian fiscal policies, as some of its opponents have alleged, this explicit acceptance that deficits may reflect cyclical factors will allow governments to cut taxes and/or increase expenditures as anti-cyclical policies require. One-off or temporary payments such as those to the Anglo bondholders are also excluded from the deficit calculation.
Much depends, of course on how the adjustments are made. The IMF, the OECD and the EU, all make estimates of the structural deficit, and the first two of these agencies both regularly publish these together with other data on government lending and borrowing. I believe all three agencies use very similar methods of estimation. (Knowing nothing about this topic I found Box 1 in the paper Tony McDonald, Yong Hong Yan, Blake Ford and David Stephan, Estimating the structural budget balance of the Australian Government in the Australian Treasury’s Economic Roundup 2010, Issue 3 (http://archive.treasury.gov.au/contentitem.asp?NavId=&ContentID=1881) very useful as an introduction. This gives references to more technical papers describing the estimating processes of each of the agencies in detail.
Potential output is derived using two-factor constant-returns-to-scale Cobb Douglas production functions with some assumed rate of growth of total factor productivity. If you are an international agency needing to make regular internationally comparable estimates of the potential output of scores of countries, this may be the best you can do, but this method of calculating potential output seems particularly inappropriate for Ireland, where the sudden arrival and departure of foreign firms and fluctuating rates of international migration can change the productive potential of the economy very quickly. The agencies can presumably also use only very approximate rules of thumb when trying to calculate how tax revenues would increase to the achievement of this potential output. The only expenditure that is held to be cyclically determined is unemployment compensation.
The inevitable imprecision of these calculations makes it unsurprising that agencies can differ quite markedly in the proportion of a given government deficit they attribute to non-structural causes. For example, averaging estimates for Ireland for 2000-2009, the IMF had an average structural balance of -4.4% of GDP; the OECD had -2.2%. In 2010, the OECD had a structural balance of 25.5% where the IMF had 9.9% (presumably reflecting different treatments of payment obligations under the banking bailout scheme.) Unfortunately I do not know of a readily available internet source of EU Commission estimates.
Since Ireland is clearly in the “excessive deficit” zone, these differences are not currently of critical importance. But eventually this position will change, and one can envisage very considerable argument about the proportion of any given deficit which is to be considered non-structural. It would be advisable for the Irish government to develop its own methods of estimating structural deficits, and to be prepared to defend them under sceptical questioning if they show lower structural deficits than Commission or IMF figures. Indeed it would be worth seeking the help of one or two economics of unimpeachable international reputation, and who have not previously taken public positions in rejecting EU stabilisation policies, to guide the process of making the estimates and certifying their credibility.
There is no reason why built-in stabilisers need be confined to payments to the unemployed. A perfectly feasible Keynesian measure would be legislation that inversely relates VAT rates to the level of unemployment. Discretionary stabilisers, such as a list of non-essential infrastructure maintenance projects that could be speedily put into operation would be another possibility, though more likely to be challenged by the EU monitors.
Although the Treaty regards national budget surpluses with favour, they have a potential deflationary impact on the Eurozone as a whole. If Keynes had been involved in designing this treaty, he would probably have attempted to have supplemented the “excessive deficit procedure” with an “excessive surplus procedure”, to have forced countries with large budget surpluses to reduce these, even at the expense of greater domestic inflation. He would of course have had as little success with the Germans as he did with the Americans when he made comparable proposals about persistent creditor countries at the Bretton Woods conference in 1944 that set up the postwar international monetary system.
The concept of a structural budget deficit has been in use for decades by the IMF, but it is not one that is in general discourse, even among economists. It does not refer to the structure of an economy, which is normally taken to be a set of productive activities, many of which are locationally fixed, and includes both capital equipment, infrastructure and human capital which can change only slowly. The EU funds designed to help areas of high unemployment to improve their productive capacity are accordingly known as structural funds. This structural deficit is quite different—it involves a snapshot (if a one-year flow measure can be called a snapshot) of how the existing system of taxes and public expenditures implies for an economy operating at full capacity.
It is easy to get confused about this difference. Finance Minister Michael Noonan interviewed on RTE’s This Week on May 20th appeared to be implying that needed structural reforms—he alluded to the retraining of unemployed construction workers—were part of the programme to reduce the structural deficit. Of course, such reforms will lead to increases in both GDP and government revenues rises (and presumably reduce the structural as well as the general deficit) but this is to use the word “structural” in its more usual meaning, rather than in the sense used in the treaty.
When the Maastricht Treaty was first agreed in 1993, there was much criticism that it required purely monetary convergence, and paid no attention to labor market convergence, either cyclical or structural. At that time unemployment rates in different Eurozone countries diverged sharply. Spain had an unemployment rate of over 20%, and Finland and Ireland rates of over 15%. In contrast, it was only 4% in Austria. There were concerns that whole countries might find themselves saddled with uncompetitive economies unable to devalue, and become permanently depressed regions within a larger federation, rather like the Italian Mezzogiorno or the US Rust Belt.
This fear appears to have been premature. The pre-crisis convergence of unemployment rates has been striking. The standard deviation of national unemployment rates was 5.1% when the treaty came into force, 3.6% in 1999, the first year of the euro and only 2.1% in 2007. Even in Spain, which appears to be the country in which unemployment has been most persistent, unemployment rates fell until in 2007 they were, at 8.3%, lower than in Germany (8.7%). The effect of the crisis has, however, been sharp divergence among national rates. The standard deviation in 2011 was 5.4%; Spanish unemployment was 21.7%.
None of this means, of course, that when stabilization has been achieved and a respectable rate of economic growth reestablished throughout the Eurozone, there will be no further need for reform. The fear of permanently depressed countries has not gone away—(see, for example a recent article by Martin Wolf (Irish Times, May 21st). This reinforces what this crisis has made sharply evident—the need for some central fiscal authority, (and also a properly empowered Central Bank). But the 1993-2007 convergence in unemployment rates suggests that such central authorities ought to be able to adopt Keynesian policies as required.