I got on a plane on Monday and flew many hours. I was in search of the public debt overhang. I read and read many articles during my journey to the other side of the planet (where these overhangs are apparently sighted on a daily basis). But in the cold (very) early (very) hours of today, I concluded my mission was a failure. The rumours of a public debt overhang, whatever that might be, remain just that – the mumbling gossip that passes for truth once the public start spreading it. In the world of facts, such an overhang eludes specification. In January 2013, at the annual American Economic Association meeting (January 4-6, 2013) in San Diego, there was a panel session with a number of allegedly “leading” economists. Their deliberations were apparently public endorsement of the claim that government debt had reached a dangerous overhang and would undermine growth prospects for the future. The policy options were limited and all involved harsh fiscal austerity – or in IMF speak “growth friendly fiscal consolidation” (which is my nomination for the joke phrase of 2013). The problem was that a few months later the IMF released a major update (October WEO) where they appeared to deny the presence of a “tipping point” – some dangerous threshold that public debt should not exceed (R&R-style). So here is how it all unfolded …
The IMF report on the panel discussion started as such:
Policymakers in advanced economies will have to resolve the problem of high government debt or they may face low growth prospects, a panel of economists said.
Which economists were the panel? None other than the Chief economist at the IMF, Olivier Blanchard, former IMF chief economist, Simon Johnson, Kenneth Rogoff (pronounced for non-English speakers “Rogue-off”!) and Thomas Sargent (whose macroeconomics text book in the late 1970s, which every body raved about has barely a sentence pertaining to macroeconomics in it).
A good group indeed.
But one that seems to have group amnesia.
In Chapter 3 of the IMF’s October 2012 World Economic Outlook – < a href="http://www.imf.org/external/pubs/ft/weo/2012/02/pdf/c3.pdf">The Good, the Bad, and the Ugly: 100 Years of Dealing with Public Debt Overhangs – we read:
Our analysis is not meant to dispute the notion that, all else equal, higher levels of debt may lead to higher real interest rates. Rather it highlights that there is no simple relationship between debt and growth. In fact, our subsequent analysis emphasizes that there are many factors that matter for a country’s growth and debt performance. Moreover, there is no single threshold for debt ratios that can delineate the “bad” from the “good.”
Note the use of the qualifier “may” in the first sentence. That is IMF code for they would love to say does but know that the empirical evidence doesn’t actually support the conclusion and when research papers find such a relationship the causality issue predicates against the – high debt causes slower growth – hypothesis.
They admit as much in the next sentences in the quote.
And the fact is that when evidence is found
Thomas Sargent distinguished himself at the meeting with this offering:
You can’t simultaneously promise that you’re going to have a low-tax state and a high government expenditure state.
He was claiming the current (assumed) debt crisis (overhang) was due to “incompatible promises that have been made to people, or incompatible understandings”.
Overhang is now the fancy term the IMF and others are using, even though it is unclear always what the overhang is in relation to – although lurking always is the “tipping point” concept from the Excel spreadsheet wizards.
The reality is that you can simultaneously promise a “low-tax state” and a “high government expenditure state” if:
1. The spending behaviour of the external sector (net exports) and the private domestic sector are consistently delivering low rates of economic growth and a mass of idle resources including labour.
2. That the net public spending is consistent with the spending gap left by 1 and it was decided that the current level of taxation (assessed at the benchmark full employment state) was equitable and placed enough purchasing power potential in the private sector.
Then you might find it necessary to increase the proportion of total GDP accounted for by public spending and reduce the proportion accounted for by taxation revenue.
There is no necessary link between cutting tax revenue and “funding” government spending. It is all about the desired public/private mix in final demand (that is, how much command on total real resources you want the public sector to have) and regulating private purchasing power (via taxes) to achieve that political desire.
One would conclude that the “incompatible understandings” were held by the Nobel (not) prize winner (the not being recognition that the economist’s prize is not a real Nobel prize.
On the topic of “public debt overhangs”, R&R were joined by a spouse to produce a paper for the Journal of Economic Perspectives in 2012 – Public Debt Overhangs: Advanced- Economy Episodes Since 1800.
This paper is now being quoted by them instead of their 2010 Excel Spreadsheet paper to validate their on-going crusade against public debt. It also contains a lot of the material that I reviewed last week in this blog – More worn out ideological prattle from R&R.
With a title like that, one would assume that we might be close to finding out what they mean by a public debt overhang. That is, I am seeking the elusive benchmark from which we measure current debt positions and then conclude whether they are hanging over, or whatever.
I wouldn’t read the paper if I was you and desired 30 minutes of constructive engagement with something today. I have read it several times, which just goes to show you that: (a) I am an economist; and (b) I live a dull professional existence.
Anyway, cutting to the chase – we don’t learn anything. We see graphs going up to the sky and lots of talk about the graphs but the closest we get to a benchmark is this statement:
… exceptionally high public debt, defifined as episodes where public debt to GDP exceeded 90 percent for at least five years.
They also refer to this as “90 percent as the critical threshold, as in Reinhart and Rogoff 2010a, b” – referring to their Excel spreadsheet paper as some sort of robust piece of analysis, when as we now know, it really should be a demonstration case for first-year IT students on how not to demonstrate incompetency at a relative easy piece of software.
We obviously cannot conclude conclusively as many others have done that the way they sorted the data and left out certain rows/columns of numbers and countries from their calculations was all the difference between “proving” their desired result and not, was an example of academic fraud.
And once you include the full sample the results do not support their strong conclusion.
No, we cannot conclude anything from that – conclusively.
The point is that there is no benchmark specified but that doesn’t stop them raving on throughout the paper about public debt overhangs.
Whoever refereed this paper – if anyone – did a very bad job. It does shine a very good light on the JEP. But then we knew the JEP was just another one of those rags that perpetuate mainstream lies.
Anyway, some economists have been digging further into the “tipping point” conjecture (assertion).
There was a nice summary Op Ed –
Public debt and economic growth: There is no ‘tipping point’ – which has analysed all the recent literature on the relationship between public debt and economic growth.
The paper was motivated by the fact that the “idea that there is a common tipping point … is still widespread”. As noted above, this is largely due to the R&R promotional tours that seem to be going on despite being exposed for their Excel spreadsheet chicanery.
A related paper – Public Debt and Economic Growth in Advanced Economies: A survey – co-authored by one of the Voxeu Op Ed contributors provides back-up evidence to the Op Ed.
They note that even one of the earlier benchmark studies (Elmendorf and Mankiw, 1999) in the public debt/growth literature shows that the alleged impact on growth of a rise in the public debt ratio is tiny. For example:
… our calculations indicate that increasing the debt by 100 per cent of GDP would reduce annual GDP growth by approximately 20 basis points in the first twenty years.
So would a 20 basis points (that is, 0.2 of one percent) impact spread over two decades justify harsh fiscal austerity and the rise and maintenance of mass unemployment that have been going on in the name of “fiscal consolidation”.
I am not quoted Elmendorf and Mankiw’s paper as if it is sound. It is a dreadful piece of work – full of nonsensical claims about Ricardian Equilivalence and crowding out – but I don’t feel like traversing that AGAIN today (bit tired).
The point is that when these characters give it their best shot and are forced to actually calibrate their claims – put them into hard numbers – the result is a joke.
The estimated growth losses pale into miniscule insignificance when compared to the massive permanent losses they inflict by way of fiscal austerity in the name of avoiding these miniscule growth losses.
It is astounding how poorly informed the public are about what my profession gets away with.
The more detailed paper considers all the recent literature on the relationship between public debt ratios and economic growth.
They variously conclude as follows:
1. Using endogenous threshold techniques provide not clear answers – in one case, there was a negative relationship between public debt ratios of 90 to 115 per cent of GDPand growth and then a positive relationship after that! So governments should spend up bigger! In fact, this just shows there is non-linearity which most simple studies like R&R did not consider.
2. The existence of possible covariates (that is, that some other causal factor is correlated with both growth and debt) would suggest that “simple correlations discussed above may suffer from an omitted variable bias” they find that once non-linearities are taken into account, the “standard regressions or group comparisons do not show evidence of a threshold effect”. READ: No threshold effect.
3. Once endogeneity is controlled for (that is, the fact that a negative relationship betewen public debt and growth could be due to growth driving a rise in public debt via higher deficits and institutional arrangements that match deficits to public debt issuance) – “the negative correlation between debt and GDP growth vanishes”. READ: vanishes.
4. Once non-linearities are carefully modelled the overwhelming conclusion is that “there is no statistically significant relationship between debt and growth in advanced economies” READ: no relationship.
5. On the issue of what is the best public debt measure to use in studies like this they conclude: “Since net debt is hard to compute and rarely comparable across countries, most papers that study the relationship between debt and growth use gross debt, even if this measure of debt is not a good indicator of the government’s financial situation.” READ: any relationships detected might in fact be meaningless.
The overwhelming conclusion is that:
… the case for a causal effect running from high debt to low growth still needs to be made. Apart from causality issues, we also show that the evidence of a common debt threshold above which growth collapses is far from being robust
The Op Ed article mentioned above also supports that conclusion. It concludes:
In sum, the commonly found 90% debt threshold is likely to be the outcome of empirical misspecification – a pooled instead of heterogeneous model – and subsequently a misinterpretation of the results, whereby it is assumed that the pooled model estimates imply that a common nonlinearity detected applies within all countries over time.
Have you read any articles in the WSJ, Financial Times, the other daily papers that report the claims about debt overhangs, which have covered this sort of deep research?
There is a major issue in what the public actually is exposed to. Simple claims about public debt overhangs resonate because the public does not know what questions to ask and the economics journalists who should no better just reiterate the press releases put out by the “think tanks”.
If they really did their jobs, the economics journalists would be seeking to better educate the public on the deeper more detailed research that is done which shows these simple claims are just ideological claptrap.
But then the journalists would have to know how to read the technical literature themselves and in many cases that is a problem.
That is enough for today!
(c) Copyright 2013 Bill Mitchell. All Rights Reserved.