Why justifying austerity just got harder

Blunders in the Reinhart-Rogoff theory


The Internet exploded this week with the revelation that researchers from the University of Massachusetts had debunked the widely-accepted thesis from Harvard professors Carmen Reinhart and Kenneth Rogoff that a country’s economic growth could be pegged to the size of its government debt and that an economy would start to go off a cliff when a country’s debt-to-GDP ratio rose above 90 per cent.

The Massachusetts economists, Thomas Herndon, Michael Ash, and Robert Pollin, crunched data given to them by Reinhart and Rogoff for 20 countries and years from 1946 to 2010 and found that the economists had made a number of calculation errors, namely excluding several years when some countries had both high debt and high growth in GDP (mostly in the years after the Second World War.)

They also found that five of the 20 countries cited in the study had accidentally been dropped from the analysis because of a “spreadsheet error” that essentially involved Reinhart and Rogoff skipping the first five countries of the alphabet.

Correcting for the mistakes, the Massachusetts researchers found that annual GDP growth was, on average, 2.2 per cent for countries with debt-to-GDP ratios above 90 per cent. That is in stark contrast to Reinhart and Rogoff’s findings that GDP would instead fall by -0.1 per cent.

Reinhart and Rogoff are now being skewered by their critics, particularly because politicians (including our own) have repeatedly cited the 90-per-cent threshold when justifying austerity budgets and the mistakes open up the possibility that maybe deep cuts aren’t needed to kick start an economy after all. (“How Much Unemployment did Reinhart and Rogoff’s arithmetic mistake cause,” is how The Guardian newspaper put it.) Reinhart and Rogoff have apologized for the errors, but insist they don’t alter their central conclusions that more debt is bad for the economy.

More important for Canadians is the fact that Canada was one of the countries omitted from Reinhart and Rogoff’s analysis, thanks to being pretty high up in the alphabet. (The others were Australia, Austria, Belgium, and Denmark)

In fact, while the Massachusetts economists found that some countries (namely the U.S.) fit Reinhart and Rogoff’s model quite nicely, Canada doesn’t. Their analysis of 64 years of Canadian economic data suggests a fairly weak link between debt levels and GDP growth in this country. If anything, Canada’s economy actually grew with higher levels of public debt. It slowed when debt topped 90 per cent of GDP, but average growth in those years was still 3 per cent of GDP a year, well above the 2.2 average for the 20 countries in the study.

Here is how the Massachusetts researchers calculated the relationship between debt and GDP growth in Canada and other countries that didn’t fit the Reinhart-Rogoff model:

Note: Spain, Norway and Finland have never topped the 90 per cent debt-to-GDP ratio.

Canada’s “sweet spot” for public debt, according to the Massachusetts analysis, is somewhere between 60 and 90 per cent of GDP, which is precisely where Canada stands today (granted, at the high end of that range, at around 86 per cent). Even so, they found that  between 1946 and 2010, Canada had 45 years where the public debt was low, but GDP growth was below optimum levels, which raises the question of whether Canada’s debt was for many years too low.

This analysis is worth considering given the federal government’s austerity plans and the IMF’s recent warning that the government shouldn’t scrimp and save its way into a recession. The Massachusetts economists concluded that Reinhart and Rogoff’s analysis presents “a false image that high public debt ratios inevitably entail sharp declines in GDP growth,” which should prompt countries to reassess their austerity agendas.

Then again, Parliamentary Budget Officer Kevin Page has consistently warned that the federal government’s debt calculations need to take into account rising levels of provincial debt and that even if the federal government slashed its budget back into surplus, provincial debt could swallow up 200 per cent of GDP by 2080.

While the Massachusetts analysis has poked holes in the previously unassailable relationship between debt and economic growth, even the fiercest austerity critics would acknowledge that there’s probably some threshold where too much government debt becomes a bad thing and that it’s most likely somewhere between Reinhart-Rogoff’s 90 per cent and Kevin Page’s 200 per cent projections.


Why justifying austerity just got harder

  1. I’ve heard several reports referencing the deletion of data from the immediate post-WW2 years, but little discussion of the rationale. I’m guessing that the data were excluded because they are attributable to fairly unusual circumstances – a shift from a war-driven economy to a civilian-driven economy, combined with the fact that Canada had been left relatively unscathed by the war. That combination of high debt and high economic growth is truly anomalous.

    • Nonsense. Debt stocks of varying maturities and terms should not be used as a benchmark against which to measure the flow of economic activity taking place in an economy over the arbitrary parameter of a year. To quote Robert Shiller:

      “Could it be that people think that a country becomes insolvent when its debt exceeds 100% of GDP?

      That would clearly be nonsense. After all, debt (which is measured in currency units) and GDP (which is measured in currency units per unit of time) yields a ratio in units of pure time. There is nothing special about using a year as that unit. A year is the time that it takes for the earth to orbit the sun, which, except for seasonal industries like agriculture, has no particular economic significance.

      We should remember this from high school science: always pay attention to units of measurement. Get the units wrong and you are totally befuddled.

      If economists did not habitually annualize quarterly GDP data and multiply quarterly GDP by four, Greece’s debt-to-GDP ratio would be four times higher than it is now. And if they habitually decadalized GDP, multiplying the quarterly GDP numbers by 40 instead of four, Greece’s debt burden would be 15%. From the standpoint of Greece’s ability to pay, such units would be more relevant, since it doesn’t have to pay off its debts fully in one year (unless the crisis makes it impossible to refinance current debt).”

  2. There has been much fire and brimstone in economic/finance circles
    the past couple of days (especially US).
    But R&R doesn’t “cause” austerity … it just gave support to what the
    powers-that-be were going to do anyway. it’s the accepted policy that
    we put the boots to the lower orders in order to make the fragile flowers
    of finance feel better.

  3. Oops…looks like we may have shot Greece in the head a little too early…oh well, who’s next?

  4. The notion that governments must borrow money to have economic growth is insane. We borrow money so we can have good times now while leaving our children to pay for it all.

    • I think the better question is: borrow money for what?

      I don’t expect governments to pay off things like large infrastructure projects in one go. But, I would expect them to make sure that non-CAPEX costs are in line with tax revenues.

    • Why? It’s well known among accountants and even some economists that if a company doesn’t leverage its capacity for debt enough, it will likely be left behind.

  5. The first issue to point out is that there are other factors that could lead to the supposed relationship found in this research. As the saying goes, correlation is not necessarily causation. This test is impossible to do without mediating factors, since countries cannot be randomly assigned debt-to-GDP levels. In other words, it could just be that countries that had more debt tended to be more developed, and thus, grow slower overall. The result probably does hold relatively true, but it is not as if high debt can ever be proven to cause lower growth.

Sign in to comment.