As a researcher one learns to be circumspect in what one says until the results are firm and have been subjected to some serious stress testing (whatever shape that takes). This is especially the case in econometric analysis where the results can be sensitive to the variables used (data etc), the form of the estimating equation(s) deployed (called the functional form), the estimation technique used and more. If one sees the results varying significantly when variations in the research design then it is best to conduct further analysis before making any definitive statements. The IMF clearly don’t follow this rule of good professional practice. They inflict their will on nations – via bullying and cash blackmail – waving long-winded “Outlooks” or “Memorandums” with all sorts of modelling and graphs to give their ideological demands a sense of (unchallengeable) authority before they are even sure of the validity of the underlying results they use to justify their conclusions. And when they are wrong – which in this case means that millions more might be unemployed or impoverished – or more children might have died – they produce further analysis to say they were wrong but we just need to do more work. So who is going to answer for their culpability?
On page 43 of the latest IMF World Economic Outlook update (October 12, 2012) – Coping with High Debt and Sluggish Growth – we read the following statement:
More work on how fiscal multipliers depend on time and economic conditions is warranted.
This was part of the WEO Update, where the IMF spell out how inaccurate their estimates (biased towards zero) of the fiscal multipliers have been. I will come back to that later.
You will recall I mentioned a last week – in this blog – The moronic activity of the rating agencies – that the rule which is being used to force Eurozone nations into self-destructive fiscal austerity – the Stability and Growth Pact fiscal rule – was just made up within an hour to satisfy the political needs of the then French President, François Mitterrand.
In case you missed it – Le Parisien carried a story on September 28, 2012 – L’incroyable histoire de la naissance des 3% de déficit (The incredible story of the birth of the 3% deficit) which revealed that a senior Budget Ministry official in the Mitterrand government was asked to come up with the fiscal rules that would become the Stability and Growth Pact (SGP).
Le Parisien tracked the official down recently and asked him about the origins of the 3 per cent rule and he replied (translated from the French):
We came up with the 3% figure in less than an hour. It was a back of an envelope calculation, without any theoretical reflection. Mitterrand needed an easy rule that he could deploy in his discussions with ministers who kept coming into his office to demand money … We needed something simple. 3%? It was a good number that had stood the test of time, somewhat reminiscent of the Trinity.
As I explained in this blog – Tightening the SGP rules would deepen the crisis – the origins of the SGP fiscal rule were arbitrary and then, ex post, there were a number of research papers from the EU and OECD attempting to justify the rule on so-called “scientific grounds”.
For example, this OECD Economic Studies paper from 2000 – Estimating prudent budgetary margins for EU countries; A simulated SVAR model approach – attempted to explore the scope that the Maastricht Treaty provided EMU governments to respond to a negative aggregate demand shock (which would manifest as an output gap).
The 2000 OECD paper claimed that:
The imposition of such debt and deficit ceilings does not necessarily impose a binding constraint on the use of counter-cyclical fiscal policy because countries can run a structural deficit that is well below 3 per cent of GDP. How far below 3 per cent is likely to be enough to allow the government deficit to play its role as a shock absorber in times of an economic slowdown or recession? The answer to this question depends on the size of economic shocks, the sensitivity of deficits to those shocks, and the extent to which the authorities might want to go beyond automatic stabilisation.
In my 2008 book with Joan Muysken – Full Employment abandoned – we had a section – Stability and Growth Pact, neither stability nor growth – which reflected on our research over several years which challenged the sort of proposition that the OECD was using to justify the 3 per cent rule.
I have long claimed that the fiscal rules were too restrictive to cope with a negative demand shock of the proportions we have now witnessed. It is clear that the advice that the EU was getting at the time was that expected negative shocks would be in a certain range – well below what we have now seen possible.
The rules, in other words, were designed on an overly optimistic view of what was possible. When you look back on the literature that the elites wheeled out to support their case it is clear that they didn’t anticipate how fragile their system really was.
Had the Euro elites bothered to examine the research evidence at the time they would have realised that the 3 per cent rule was too restrictive. For example, a 1997 study by Buti et al. examined the behaviour of fiscal outcomes for EU economies between 1961 and 1996 and found that:
… during 50 episodes of recession or abrupt slowdown in growth… an excessive deficit would have occurred in eleven cases if the budget had initially been balanced, but in 28 cases if the starting point had been a deficit of 2 per cent of GDP – i.e. more than a doubling of the risk of running into excessive deficits. They also conclude that the risk of incurring an excessive deficit is high in the case of protracted recessions, even if the starting point is a balanced budget. The same conclusion is drawn with respect to exceptionally severe recessions in which real GDP falls by more than 2 per cent.
(Full reference – Buti,M., Franco, D and Ongena, H. (1997) ‘Budgetary policies during recessions – retrospective application of the Stability and Growth Pact to the post-war period’, European Commission, Economic Papers, No. 121).
There were other studies that supported that sort of conclusion which leads you to conject that the EU forced the Mitterand-rule onto the monetary system without considering the implications. Even then it was obvious that several member states would be caught out and forced to introduce growth-sapping pro-cyclical fiscal interventions if a reasonable size cyclical downturn occurred.
Further, given the logic of the EMU, these same nations would soon encounter the disdain of the “bond markets” and a Euro crisis would be inevitable.
Who is being held responsible for this culpability? Answer: the unemployed, pension recipients, and the rest of the citizens who are being punished for the incompetence of the elites.
Anyway, of relevance to today’s blog is the fact that some of this is linked to the estimation of output gaps. I have consistently argued that the way organisations such as the CBO, the IMF, the OECD, the OBR and the rest of the economic forecasting units attached to or used by governments to frame policy impose inherent ideological biases on the estimates of their output gaps.
The direction of the biases is to understate the size of the gaps and therefore conclude that the policy stance is overly expansionary when in fact it is likely to be contractionary.
The output gap measures published by the IMF and the OECD are almost certainly too conservative because they are typically based on concepts such as the NAIRU. Please read the following blog – The dreaded NAIRU is still about! – for more discussion on this point.
This is one of the reasons they keep delivering grossly inaccurate estimates in their pontifications.
What it means is that the estimates of the cyclical components of the actual deficits will almost assuredly biased downwards, which means the estimates of the “structural” component will be biased upwards.
This, in turn, means that we will be led to conclude that the discretionary fiscal position adopted by the government in question will be more expansionary or less contractionary than it actually is.
In the blog – Tightening the SGP rules would deepen the crisis – I did some calculations which you can read if interested.
What I showed was that the elites supported by the OECD etc claimed that even if a nation was running structural deficits around 1.5 per cent of GDP they would have more than enough space to avoid breaching the SGP threshold of 3 per cent based upon what the OECD was claiming were the “mean value of the maximum output gaps recorded in recessions in the major EU economies during 1975-1997”.
The problem was that these estimates of “fiscal space” were grossly overestimated.
First, the OECD and others typically assumed that before the recession all nations were at full employment already (zero output gap) which is clearly not a viable starting point. The output gaps were persistent across the Eurozone in many economies so they went into the crisis with cyclical deficits. Not all the deficits were of a structural nature.
When you consider that and note that some governments were running actual surpluses before the crisis then you start to appreciate how much fiscal drag was in the system and how reliant on private credit growth the region became. That was an unsustainable but typically neo-liberal growth strategy.
Further, the austerity approach en masse relies on the private sector offsetting this fiscal drag (and the drag on exports from the austerity given how closed the Eurozone is as a trading market). In the same way that the UK austerity assumes an already heavily indebted private sector will accumulate more debt to get the economy growing the same logic applies to the Eurozone. That is, they are just repeating the errors that led to the crisis.
Second, the analysis ignores problems of path dependence. For example, a nation could be caught in a sequence of recessions which would, under current arrangements, ratchet the public debt ratio up over time. So each recession would reduce the “space” for structural deficits in the future under the SGP rules, irrespective of whether the nation recovered to full employment in between the episodes.
Some cursory empirical analysis (in the blog mentioned) showed that even using the OECD output gaps (which were biased downwards) the current recession generated output gaps which are considerably larger than those that we considered to be reasonable to justify the SGP restrictions.
This means that there was no chance that the majority of budgets in the EMU could withstand the current crisis and remain within the SGP rules, even if they started from balanced positions.
Thus, it becomes obvious that: (a) the starting pre-crisis fiscal position for many nations was too contractionary (hence the persistently high unemployment even when the economies were growing; and (b) the resulting increase in the fiscal deficits were largely cyclical as a result of the dramatic increase in the output gaps that the collapse in private spending generated.
So the SGP limit could not even cope with a major cyclical swing. Which means that using it as a guide for the conduct of discretionary fisal policy changes is disastrous.
And so it has been. We now have nations undergoing vicious and harsh pro-cyclical fiscal policy shifts (that is, austerity) and being threatened with excessive deficit fines – when their deficit positions are just signalling how large the output gap is.
The pro-cyclical fiscal policies that are being forced on these nations, in turn, make the output gap worse and invoke even harsher penalties from the Troika. The governments are then forced to make further cuts in discretionary net public spending and the cyclical budget component gets worse.
It is almost certainly the case that the cyclical component alone breaks the SGP constraints – for Finland, Greece, Ireland, Netherlands, Spain and probably Italy.
The question that arises is – what sort of fiscal rule would force a nation to run a budget surplus at full employment as a matter of course (irrespective of the other sectoral balances) – if the cyclical response that might accompany a severe recession would violate that rule?
So the made-up 3 per cent rule is being used to violate even the most basic concept of responsible fiscal management. Fiscal policy should not be used to kill growth and drive up unemployment when there is no inflation threat and growth is collapsing as a result of inadequate private spending.
This is how arbitrary the neo-liberal period has been. Yet these arbitrary rules and assessments – all part of an elaborate smokescreen or charade – cause real damage to the lives of working people and rarely undermine the capacities of the elites. There are countless high paid officials in Brussels strutting around making all sorts of statements about the need for nations to cut welfare, wages, jobs and the like based on a rule that was just made up on the spot without any economic justification or authority.
But it gets worse.
First, there was a publication from the UK-based Capital Economics which came out last week and was reported in the FT Alphaville article (October 4, 2012) – Why the UK output gap could be a chasm. As far as I can tell this is a subscription report from Capital Economics who seem to charge enormous amounts for clients to learn the obvious. Good marketing I suppose. [EDIT: After I had published this blog, Capital Economics kindly sent me the relevant report. I appreciated that – and it is a very interesting analysis].
The article tells us that Capital Economics has concluded that the output gaps currently being used by policy makers in Britain are significantly understating the true gap between potential GDP and actual GDP.
The policies that are being based on the official output gap estimates thus assume that the UK economy has less excess capacity than is the true case. This means that the policies introduced will be almost certainly more contractionary than is being assumed.
This is one of the reasons why the official estimates from organisations like the IMF are so inaccurate (they are always overly optimistic about real output and downplay unemployment rises).
The following graph is reproduced from the Capital Economics Report.
So you can see that at present, real output is some 14 per cent below a trend-extrapolated growth rate (based on 2.5 per cent growth). Beware though. The extrapolation – which is reminiscent of the so-called peak-to-peak method of linear extrapolation which we used to use to come up with measures of potential output – assume that the starting point (a peak) – is a full capacity point.
If the the first-quarter 2001 Real GDP estimate contained some excess capacity (that is, the economy was not at full employment) then the true gap would be larger than 13.7 per cent, by the size of the excess capacity in that quarter.
The Capital Economics Report said that most forecasters claimed in relation to 2007:
That the economy was overheating in that year is a view now also held by the major international forecasters, including the IMF, OECD and European Commission
But it is likely that these estimates were “generous” especially when one considers other evidence which does not support a fully employed, overheating economy.
But that aside, Capital Economics say that bodies such as the UK Office of Budget Responsibility (OBR), which provides the modelling for Treasury, have deliberately assumed that the recession has destroyed a massive amount of productive capacity.
In relation to the above graph, this would mean that the upper (extrapolated) line would be significantly below the 2.5 per cent extrapolation. The Capital Economics report says that:
By far the bigger factor is the view that the recession caused a significant permanent loss of productive potential. The OBR’s output gap forecast in 2012 (2.6%), its estimate of the degree to which the economy was overheating in 2007 (2%) and the extent to which GDP is currently falling short of its pre-crisis trend (14%), implies a permanent loss of output of something like 9% of GDP.
The upshot is that Capital Economics believe the idea that the potential output has fallen that far is far-fetched and contrived to deliver a politically palatable scenario.
The OBR-types claim that if the output gap was larger than their 2.6 per cent (and certainly up to 14 per cent) then the UK should be deflating rapidly.
This is the same sort of arguments that came up in the mid-1970s when stagflation became a problem. The Friedman Monetarists attacked the then Keynesian orthodoxy on the grounds that demand-management had failed to deliver because economies were being confronted with rising inflation at a time when there was rising unemployment.
This was interpreted as meaning there was no coherent trade-off between the two evils and that the full employment unemployment rate (the NAIRU) was almost certainly higher than believed (that is, in the parlance of today’s blog – the output gaps were low).
But the inflation at present in the UK is either policy-driven (VAT increases) or supply-driven (energy prices etc). Demand-pull inflation – where the output gap is low and nominal demand growth strains the capacity of the economy to respond in real terms (increasing output) is non-existent at present.
Capital Economics concludes that:
… for the sake of argument, we don’t think that it is out of the question that the economy was close to its potential in 2007 and that very little output has been permanently lost in the recession. On that basis, the output gap would be about 13% of GDP.
They suggest that a reasonable conservative estimate of the output gap would “still be about 6% of GDP”. which is a far cry from the official estimate of 2.6 per cent.
Which means, following on from the earlier discussion about the SGP, that current fiscal and monetary policy settings in the UK are excessively contractionary and are likely to cause more damage than is being predicted.
Further, it means that there is undoubtedly significant scope for fiscal expansion to drive nominal aggregate demand closer to capacity without any threats of demand-pull inflation.
Finally, back to the IMF. Their admissions about fiscal multipliers is the other part of the equation.
The underestimation of output gaps is one thing. But the underestimation of the size of the fiscal multiplier is another. The first error leads to imposing policies that are excessively contractionary.
The second error then leads the government to underestimate the damage that the fiscal contraction causes. Both lead to disastrous outcomes accompanying fiscal austerity and explain why the world is in such a mess.
We are in the hands of drunken sailors who have think the rocks are further away than they are and who don’t know how strong the wind is that is pushing them towards the reef. Even as they sink – they chant a happy tune.
On Page 41 of the IMF WEO Update we encounter “Box 1.1. Are We Underestimating Short-term Fiscal Multipliers?” There is a lot of technical talk contained in the box, which is meant to give the impression that this is science we are dealing with and the IMF modellers are on top of their game.
Nothing could be further from the truth.
Here is the takeway in their own words:
he main finding, based on data for 28 economies, is that the multipliers used in generat- ing growth forecasts have been systematically too low since the start of the Great Recession, by 0.4 to 1.2, depending on the forecast source and the specifics of the estimation approach. Informal evidence suggests that the multipliers implicitly used to generate these forecasts are about 0.5. So actual multipliers may be higher, in the range of 0.9 to 1.7.
What does that mean in English? Well lets say, as part of their austerity mania, that the government cuts spending by $1 billion. Once all the adjustments are exhausted the official estimates would suggest that real GDP falls by $500 million (based on a multiplier of 0.5).
However if the multiplier was 1.7, real GDP would fall by $1.7 billion – more than 3 times the official estimate.
In other words, the estimates of the damage caused by the IMFs constant hectoring for governments to impose fiscal austerity, have been grossly understated.
I have run out of time today – but consider the human casualties of all of these “errors”, which are systematically being made by my profession.
Economists strut around with these papers and quote all sorts of technical gobblegodook but in the end they haven’t a clue what they are talking about. Their jobs, however, are not the ones that are risk.
In turn, the ideologues in the EU and the British government and all the rest of them use the estimates provided by these incompetent economists to impose harsh, job-destroying policies on their populations.
And then we are told:
More work on how fiscal multipliers depend on time and eco- nomic conditions is warranted.
So all of this is being imposed on people based on research evidence that doesn’t warrant the status of being called knowledge. Now we know how wrong all these agencies and government departments have been.
So who is going to be held responsible for the damage that is being caused?
That is enough for today!
(c) Copyright 2012 Bill Mitchell. All Rights Reserved.