For most people -- arguably even for most economists -- all that mattered was the knowledge that countries with debt over 90 percent of GDP tend to have slower growth than countries with debt below 90 percent of GDP. Simple and clear. Everyone who received that message recognised it for what it was. In the cloudy consciousness of humble non-statisticians, it was a vital warning on debt issued by two cautious and meticulous researchers - Reinhart and Rogoff.
Public availability of that knowledge exerted pressure upon policymakers to recognise adequately the risks of large debt. The Reinhart-Rogoff data planted seeds of caution in the minds of policymakers under ideological siege from Keynesians clamouring for additional public debit and deficit spending during slow growth.
The discovery of a relatively insignificant “coding error” in the Reinhart-Rogoff data analysis aroused the Keynesian ‘Fourth International’ wolves. They seized on it as a sign of weakness, they circled, and they struck with lighting speed, salivating, and snarling feigned outrage.
Armed with the ‘coding error’ these Debt ‘n’ Deficit Defenders sought to undermine warnings about dangerous debt-to-growth correlations. Joseph Schumpeter may have been the first to notice that a natural constituency of the Keynesians are the anti-capitalist ivory tower intellectuals. This gives the Keynesians great influence in mass media and the academy, and, by extension, a disproportionate hold over public opinion. Luckily, key policymakers, politicians, and central bank chiefs were not fully swept up by Keynesian passion. Chaps like Krugman are angry about that. Right thinking people, however, will feel justifiably aggrieved that pseudo-scandalising about Reinhart-Rogoff data absorbed so much time and energy, so much scarce ‘public space’.
I am pleased to see that this time-wasting could be coming to an end. Yesterday the Keynesian kingmaker, wolf pack leader, and chief ideologue, Paul Krugman, ungraciously climbed down. In response to Carmen Reinhart’s polite but insistent and legitimately outraged letter, Paul Krugman conceded the following statement is TRUE -- “countries with debt over 90 percent of GDP tend to have slower growth than countries with debt below 90 percent of GDP”.
(Not coincidentally this is the statement that I attributed - above - to ordinary folk drawing common sense conclusions from the pre-coding-error Reinhart-Rogoff data).
Krugman goes on to say another statement is UNTRUE -- “growth drops off sharply when debt exceeds 90 percent of GDP”. Well … a high school graduate would see straight away that in these historical data sets the ‘threshold’ is intended as a guide to potential cause-or-effect, and does not mean a cliff which you drop over suddenly and helplessly (it's not like the getting married threshold). Think of it, hypothetically, this way -- a person might be told of a 'statistical' threshold of 60 years of age beyond which brain functions decline. Krugman surely does not think any reasonable person would interpret the threshold as precipitous.
Did Reinhart-Rogoff even say this? I doubt it. I can’t find it. They do use the word “sharply”. Here’s a Reinhart-Rogoff statement with the word “sharply”. It’s in the famous paper ‘Growth in A Time of Debt’ whose ultimately trivial ‘coding error’ sparked the Keynesian pseudo-scandal:
“A general result of our debt intolerance analysis [is] that as debt levels rise towards historical limits, risk premia begin to rise sharply, facing highly indebted governments with difficult tradeoffs. Even countries that are committed to fully repaying their debts are forced to dramatically tighten fiscal policy in order to appear credible to investors and reduce risk premia. The link between indebtedness and the level and volatility of sovereign risk premia is ripe for revisiting in light of the comprehensive cross-country data on government debt.”
Words of wisdom. Things that can move “sharply” due to debt are perceptions of risk.
That’s the essential observation. I looked at this behavioural dimension of sovereign-level risk in a previous post. I pointed out that debt-to-GDP data don’t claim to tell us about why debt might damage growth prospects. Excessive debt, I concluded, can cause individuals, firms, and governments to change their behaviour in risky and self-defeating ways. My fellow contributor Michael Pettis seems to agree. He offers some great examples.
The manufactured pseudo-scandal over Reinhart-Rogoff’s trivial coding error ignores this question of why and how debt slows growth. In fact, Krugman, in his ungracious climb down yesterday, said a most strange thing. He accused Reinhart-Rogoff of “downplaying causality issues”. Of course they did! They are working economists in the field, not former-trade-theorists-turned-newspaper-columnists and self-confessed ideologues.
Reinhart-Rogoff know they cannot pinpoint causality in debt-to-GDP ratios. It’s a judgement call based on weight of evidence and theory accumulated over many years from many sources. In this judgement call I trust the try-genuinely-to-be-objective-and-non-political authors of ‘This Time Is Different’, the book summarising the most comprehensive comparative historical study of its kind. Based on that study (and for the same reasons outlined in my earlier post and the piece by Pettis) they might justifiably lean toward a view that the Truly Big Issue is causality from debt to growth. Yet Rogoff is on record as saying the causality could run both ways.
“Everyone has an ideology … So yes, I’m an ideologue … Let’s not pretend we share more than we do” said Krugman. I agree 100%. I respect an ideologue who is proud to wear his ideology well. Tiresome and rude as he becomes, Krugman has in the past worn the vulgar Keynesian ideology in entertaining fashion. It’s a relevant social role he has there.
But when a thought-leader is wrong, when he is tacitly admitting to having been wrong, and after having wasted so much of our collective reading time in the past 5 weeks with the hysteria about Reinhart & Rogoff, Paul could now be a little more gracious. To be frank, Paul owes Carmen and Ken an apology. He should grovel for it.