Saturday, October 05, 2013

How important was the “structural balance” screw-up in driving European austerity?


I’m really glad to see that this European screw-up is, eventually, making headlines and that the European Commission is reconsidering the operational details of its production function method to estimate potential output and structural deficits (see WSJ).
As reported by the WSJ’s Real Time Brussels blog, the issue has become important, as the new European Fiscal Compact, which entered into force on 1 January 2013, requires that the structural deficit for euro-area Member States be less than 0.5%. So since “the European Commission uses [the structural balance] metric — the actual government budget balance adjusted for the strength of the economy – to determine how much austerity is needed; getting it wrong has big consequences.”
The question is whether “getting the structural balance wrong” in 2010 – the time at which Europe started to become obsessed with fiscal austerity – mattered in driving the amount of consolidation.
I first came across and pointed out the weakness of the structural balance calculations of the European Commission in 2010 in a series of Goldman Sachs publications that I then summarized with my co-author Natacha Valla on VoxEU. The basic storyline was and remains simple. The European Commission drastically revised downward its estimates of potential GDP as the crisis hit, which automatically increased its structural deficit measures for European countries.
This raised 2 questions?
1/ Did it make sense? 2/ And did the downward revisions to potential GDP matter for the size of consolidation packages?
First, did the downward revisions to potential GDP – that the European Commission introduced early in the crisis – make sense?
The European Commission uses a production function methodology for calculating potential growth rates and output gaps (see here). It features a simple Cobb Douglas specification where potential output depends on TFP and a combination of factor inputs (potential labor and capital). Importantly for what follows, potential labor input is calculated as:
Working age Population x Participation rate x Average hours worked x (1 - NAWRU)
It’s important to focus on potential labor input since the bulk of the revisions applied between 2008 and 2010 to potential GDP arose because of revisions to labor input.
Figure 1. Decomposition of the revisions (2010 vs. 2008 vintage) to potential GDP: Spain
Source: VoxEU. Note: The vertical axis is in percentage points.
And as you can guess from the drawing by Manu Cartoons, a large part of that decrease came from an increase in the Commission’s estimate of structural unemployment: the now infamous NAWRU. The European Commission uses the non-accelerating wage rate of unemployment (NAWRU) as an estimate for structural unemployment. We can discuss the pros and cons of NAWRU as a dynamic measure of structural unemployment in normal times, but the following graph should basically scream at you that this measure has failed at separating cyclical and structural unemployment during these extraordinary times.
Figure 2: Actual unemployment and NAWRU: Spain
Source: European Commission, 2013 spring forecast exercise
So although one can argue that a crisis can temporarily – as workers need time to adjust to the new sectoral and geographical composition of jobs – or permanently – because of hysteresis effects – decrease potential labor input, the size of the adjustments applied by the European Commission appears clearly inadequate.
Second, did it matter and was it more important than the fiscal multiplier screw-up in driving austerity?
As pointed out by Real Time Brussels, a reconsideration of the methods that the European Commission uses for estimating potential output could now cut the estimated structural deficits of the periphery countries and mean less austerity given the 0.5% rule of the Fiscal Compact. But I don’t think that, back in 2010, the structural deficit numbers played an important role for the countries that were under the heavy pressure of bond markets as they mostly focused on the nominal deficit and debt to GDP ratio metrics in designing their consolidation packages.
The structural balance screw-up may have mattered, however, for Germany (and for other core countries to the extent that their fiscal strategy mostly mimicked that of Germany) as it designed its consolidation plan for the 2011-2016 period based on the structural balance metric. In my VoxEU piece, you can see that the downward revision to potential GDP applied to core countries was not negligible. For Germany, the European Commission revised the potential GDP growth rate by about 0.5 percentage points.
I need to find the output gap and the cyclical-adjustment parameters used both in 2008 and 2010 by the European Commission to quantify by how much austerity would have decreased had Germany followed the same consolidation path (similar speed and end-points in terms of structural deficit) but used the 2008 vintage version of its potential output estimates rather the revised 2010 estimates. But my guess is that the consolidation efforts Germany planned for were to a significant extent unnecessary restrictive, even taking as given its objective of converging to a structural deficit of 0.35% by 2016.
Given the sheer size of core countries for the euro area, this operational mess-up may have been quite significant in delivering an inappropriate overall fiscal stance and hindering rebalancing. As a matter of fact, I wonder if this operational mess-up wasn’t more important than the fiscal multiplier screw-up in delivering this outcome. Granted, fiscal consolidation would have been less drastic in the periphery without the fiscal multiplier screw-up. But the biggest European fiscal policy failure wasn’t so much that of the periphery, which had pretty much no choice than drastically consolidate in the absence of a proper lender of last resort. The biggest fiscal policy mistake was the early consolidation efforts of the core, which started - and were amplified by the structural balance screw-up - in 2010 although in the absence of default risk, debt adjustment should be very gradual”.
Jeremie Cohen-Setton (@JCSBruegel) is a PhD candidate in economics at UC Berkeley and an Affiliate Fellow at Bruegel. He specializes in Macroeconomic Policies and Macroeconomic History and worked previously as an economist at HM Treasury and at Goldman Sachs. Jeremie blogs at ecbwatchers.org and is the main author of the blogs review at bruegel.org.

8 comments:

  1. Anonymous5:47 AM

    The real mess up was the imposition of the SGP itself. It was based on a complete misreading of the cause of the Euro crisis, either due to stupidity or deliberately, i.e. that we are all Greece. The rest is secondary, it;s just a corollary.

    ReplyDelete
    Replies
    1. I'm not sure I get your point. The SGP predates the crisis. Maybe you're referring to the Fiscal Compact, but as you can see in my post the medium term fiscal plan of Germany was designed before the Fiscal Compact was implemented.

      Delete
  2. Anonymous7:09 AM

    Hi Jeremie, as someone with first-hand knowledge of what is happening within the EU's output gap working group, I must tell you that unfortunately there is in reality absolutely no backtracking currently going on re the current nawru and potential output methodology... Regards!

    ReplyDelete
    Replies
    1. I had indeed missed this follow-up report by Matthew Dalton http://online.wsj.com/article/SB10001424052702304213904579095481787193644.html. This is very concerning.

      Delete
    2. Anonymous5:17 AM

      Sorry for the anonymity, but my position doesn't allow me to speak out freely...

      Delete
    3. Jeremie this is an important enough consideration that you should make an appended note on your original post

      Delete
  3. Anonymous11:47 AM

    Jeremie, you might want to take a look at this regarding your question on German fiscal consolidation: www.intereconomics.eu/archive/year/2013/1/843/
    Best regards
    EK

    ReplyDelete
    Replies
    1. Thanks. Well written piece indeed.

      Delete