Yesterday I commented on the intriguing box in which the IMF staff challenges one of the tenets of the Washington consensus, the link between labour market reform and economic performance.
But the IMF is not new to these reassessments. In fact over the past three years research coming from the fund has increasingly challenged the orthodoxy that still shapes European policy making:
- First, there was the widely discussed mea culpa in the October 2012 World Economic Outlook, when the IMF staff basically disavowed their own previous estimates of the size of multipliers, and in doing so they certified that austerity could not, and would not work (of course this led EU leaders to immediately rush to do more of the same).
- Then, the Fund tackled the issue of income inequality, and broke another taboo, i.e. the dichotomy between fairness and efficiency. Turns out that unequal societies tend to perform less well, and IMF staff research reached the same conclusion. And once the gates opened, it did not stop. The paper by Berg and Ostry was widely read. Then we had Ball et al on the distributional effects of fiscal consolidation (surprise, it increases inequality). Another paper investigated the channels for this link, highlighting how consolidation leads to increased inequality mostly via unemployment. And just last week I assisted to a presentation by IMF economists showing how austerity and inequality are positively related with political instability.
- On labour markets, before yesterday’s box 3.5, the Fund had disseminated research linking increased inequality with the decline in unionization.
- Then, of course, the “public Investment is a free lunch” chapter three of the World Economic Outlook, in the fall 2014.
- In between, they demolished another building block of the Washington Consensus: free capital movements may sometimes be destabilizing…
These results are not surprising per se. All of these issues are highly controversial, so it is obvious that research does not find unequivocal support for a particular view. All the more so if that view, like the Washington Consensus, is pretty much an ideological construction. Yet, the fact that research coming from the center of the empire acknowledges that the world is complex, and interactions among agents goes well beyond the working of efficient markets, is in my opinion quite something.
What does this mass (yes, now it can be called a mass) of work tells us? Three things, I would say. First, fiscal policy is back. it really is. The Washington Consensus does not exist anymore, at least in Washington. Be it because the multipliers are large, or because it has an impact on income distribution (and on economic efficiency); or again because public investment boosts growth, fiscal policy has a role to play both in dampening business cycle fluctuations and in facilitating stable and balanced long term growth. The fact that a large institution like the IMF has lent its support to this revival of consideration for fiscal policy, makes me hope that discussions about macroeconomic policy will be less ideological, even once the crisis will have passed.
The second thing I learn is that the IMF research department proves to be populated of true researchers, who continuously challenge and test their own views, and are not afraid of u-turns if their own research dictates them. I am sure it has always been the case. What is different from the past is that now they have a chief economist who seems more interested in understanding where the world goes than in preaching a doctrine.
The third remark is more problematic. If I write a paper saying that austerity will not be costly because multipliers are 0.5, and 2 years later retract my previous statement and argue that austerity is in fact self defeating, the impact on the world is zero. If the IMF does the same, during the two years huge suffering will be needlessly inflicted to masses of people. This poses a problem, as research by definition may be falsified. In the past an institution like the IMF would never have admitted a mistake. And we certainly do not want to go back there. Today they do admit the mistakes, but the suffering remains. The only way out to this problem is that the “new” IMF should learn to be cautious in its policy prescriptions, and always remember that any policy recommendation is bound to be sooner or later proven inappropriate by new data and research. We don’t live in a black and white world. Adopting a more prudent stance in dictating policies would be wise (in Brussels as well, it goes without saying). And of course, the disconnect between the army and the general is also a problem.
I am ready to bet that the latest IMF World Economic Outlook, that was presented today in Washington, will make a certain buzz for a box. It is box 3.5, at page 36 of chapter 3, which has been available on the website for a few days now. In that box, the IMF staff presents
lack of evidence on the relationship between structural reforms and total factor productivity, the proxy for long term growth and competitiveness. (Interestingly enough people at the IMF tend to put their most controversial findings in boxes, as if they wanted to bind them).
What is certainly going to stir controversy is the finding that while long term growth is negatively affected by product market regulation, excessive labour market regulation does not hamper long term performance.
It is not the first time that the IMF surprises us with interesting analysis that goes against its own previous conventional wisdom. I will write more about this shortly. Here I just want to remark how these findings are relevant for our old continent.
imposed to embraced by eurozone crisis countries has taken the shape of expenditure cuts and labour market deregulation, whose magic effects on growth and competitiveness have been sold to reluctant and exhausted populations as the path to a bright future. I already noted, two years ago, that the short-run pain was slowly evolving into long-run pain as well, and that the gain of structural reforms was nowhere to be seen. The IMF tells us, today, that this was to be expected.
The guy who should be happy is Alexis Tsipras; he has been resisting since January pressure from his peers (and the Troika, that includes IMF staff!) to further curb labour market regulations, and recently presented a list of reforms that mostly pledges to reduce crony capitalism, tax evasion and product market rigidities. Exactly what the IMF shows to be effective in boosting growth. Of course, at the opposite, those who spent their political capital to implement labour market reforms are most probably not rejoicing at the IMF findings.
This happens in Washington. Problem is, Greece, and Europe at large, seem to be light years away from the IMF research department. We already saw, for example with the mea culpa on multipliers, that IMF staff in program countries does not necessarily read what is written at home. Let’s see whether the discussion on Greece’s reforms will mark a realignment between the Fund’s research work and the prescriptions they implement/suggest/impose on the ground.
Simon Wren-Lewis has an interesting piece on structural deficits. He has issues with Pisani-Ferry’s plea for more stable structural deficit targets for EU countries. While Pisani-Ferry has a point in invoking more certainty for EU government action, Wren-Lewis argues, rightly so, that stable targets risk creating straitjackets for countries, and that the problem is mostly in the excessively short time horizon of structural deficit targets.
The fact that both Pisani and Wren-Lewis have a point highlights what is in my opinion a structural flaw of EU fiscal governance, namely its reliance on the slippery concept of structural government deficit.
To explain this simply, the idea underlying structural deficit targets is that not all deficit were created equal. if the government runs a deficit because of adverse cyclical conditions (low growth yields lower tax revenues and larger welfare payements), this deficit is “healthy” because it supports economic activity, and bound to disappear when the economy recovers. As such, governments should not be required to target cyclical deficit, but only the structural (or cyclically adjusted) deficit, which is precisely the deficit “cleaned” of its cyclical component.
The EU fiscal rule, the Stability Pact and its hardened Fiscal Compact extension, recognizes this distinction, and imposes that governments balance their budget over the cycle, which is yet another definition of structural deficit. This may seem a sensible approach, recognizing, as I just said, that not all deficits were created equal. But in fact sensible it is not.
The problem lies precisely in the word “cleaned” I used above . How do we clean headline deficit from its cyclical component, to compute the structural deficit that should be targeted by governments? This is how we should do it: We compute “potential output”, i.e. the capacity of production of the economy. From that we can obtain the output gap, i.e. the distance of actual output from its potential level; finally, by applying an estimate of how the deficit responds to the output gap, we can clean headline deficit from its cyclical component. Simple, right? Yes, in theory. In practice, we have no way to do it in a sufficiently precise way.
Any meaningful analysis of cyclical developments, of medium term growth prospects or of the stance of fiscal and monetary policies are all predicated on either an implicit or explicit assumption concerning the rate of potential output growth. Given the importance of the concept, the measurement of potential output is the subject of contentious and sustained research interest.
All the available methods have “pros” and “cons” and none can unequivocally be declared better than the alternatives in all cases. Thus, what matters is to have a method adapted to the problem under analysis, with well defined limits and, in international comparisons, one that deals identically with all countries. (emphasis is mine)
There is nothing wrong with recognizing that potential output estimates are “contentious”. Contrary to what some Talebans persist to argue, economics is a social science, subject to all the uncertainties, mismeasurements, and ambiguities that are inherently linked to human and social interactions.
Where we have a problem is in using a contentious concept as the foundation for rules in which a zeropointsomething deviation from the target may lead to sanctions and public disapproval by the EU community, with all the potential financial market disruptions associated with it.
This makes the rule non credible, because the contentious estimate may be questioned. More importantly, it leads to what Wren-Lewis fears: countries imposing harsh sacrifices to their people that may turn out to be unwarranted when the estimate is revised.
I am not clear about what fiscal rule we should have in the EU. I actually am not even convinced that we really would need one. What is certain is that two necessary conditions for any rule to be effective, credible, and reasonable are that it is not short -termist (I rejoin Wren-Lewis), and that it is based on indicators that are quantitatively as precise as possible.
The current rule fails on both ground (and don’t get me started on how crazily complicated and arbitrary it grew over time). EU fiscal governance remains founded on sand. And of course, a serious debate on its reform is nowhere to be seen in European policy circles.