Austerity economics only works if you make an Excel formula error

A new paper called Does High Public Debt Consistently Stifle Economic Growth? A Critique of Reinhart and Rogoff by Thomas Herndon, Michael Ash, and Robert Pollin from UMass Amherst tries and fails to replicate the classic work on austerity, Carmen Reinhart and Kenneth Rogoff's 2010 Growth in a Time of Debt.

Reinhart-Rogoff is the main research cited in favor of cutting public services and spending in bad economic times. It's a big part of why the local library is shutting down, why they're kicking people out of public housing, shutting down arts programs, slashing education and public transit, and laying off public employees. It purports to show that countries with high debt-to-GDP ratios of 90 percent or more are a "threat to sustainable economic growth."

In the new Amherst paper, the authors reexamine Reinhart-Rogoff's original data and conclude that the numbers don't add up. They show that Reinhart-Rogoff cherry-picked which years of high-debt GDP they measure, that they put their thumbs on the scales with "unconventional weighting" and made a "coding error" that "entirely excludes five countries, Australia, Austria, Belgium, Canada, and Denmark." This last error -- literally the wrong formula in a spreadsheet cell -- badly skews the outcome.

Here's the tl;dr: "the average real GDP growth rate for countries carrying a public debt-to-GDP ratio of over 90 percent is actually 2.2 percent, not -0.1 percent as [Reinhart-Rogoff claim]."

Selective Exclusions. Reinhart-Rogoff use 1946-2009 as their period, with the main difference among countries being their starting year. In their data set, there are 110 years of data available for countries that have a debt/GDP over 90 percent, but they only use 96 of those years. The paper didn't disclose which years they excluded or why.

Herndon-Ash-Pollin find that they exclude Australia (1946-1950), New Zealand (1946-1949), and Canada (1946-1950). This has consequences, as these countries have high-debt and solid growth. Canada had debt-to-GDP over 90 percent during this period and 3 percent growth. New Zealand had a debt/GDP over 90 percent from 1946-1951. If you use the average growth rate across all those years it is 2.58 percent. If you only use the last year, as Reinhart-Rogoff does, it has a growth rate of -7.6 percent. That's a big difference, especially considering how they weigh the countries.

Unconventional Weighting. Reinhart-Rogoff divides country years into debt-to-GDP buckets. They then take the average real growth for each country within the buckets. So the growth rate of the 19 years that the U.K. is above 90 percent debt-to-GDP are averaged into one number. These country numbers are then averaged, equally by country, to calculate the average real GDP growth weight.

In case that didn't make sense, let's look at an example. The U.K. has 19 years (1946-1964) above 90 percent debt-to-GDP with an average 2.4 percent growth rate. New Zealand has one year in their sample above 90 percent debt-to-GDP with a growth rate of -7.6. These two numbers, 2.4 and -7.6 percent, are given equal weight in the final calculation, as they average the countries equally. Even though there are 19 times as many data points for the U.K.

Now maybe you don't want to give equal weighting to years (technical aside: Herndon-Ash-Pollin bring up serial correlation as a possibility). Perhaps you want to take episodes. But this weighting significantly reduces the average; if you weight by the number of years you find a higher growth rate above 90 percent. Reinhart-Rogoff don't discuss this methodology, either the fact that they are weighing this way or the justification for it, in their paper.

Researchers Finally Replicated Reinhart-Rogoff, and There Are Serious Problems. [Mike Konczal/Next New Deal]

Does High Public Debt Consistently Stifle Economic Growth? A Critique of Reinhart and Rogoff

(via Techdirt)


  1. Where “Excel error” means that the data are not lining up with the “correct answer”, so we need to keep fiddling with the formula until they do.

    The “Excel error” is barely relevant to the assessment of a paper that refused to publish data and was steered away from peer review.

    1. Did you even read the paper?  You seem to think that omitting the countries that start with A is good methodology, and including them is “fiddling with the formula.”  Really?

      1. Did you read it?

        There was a mistake in the Excel worksheet, but according to the actual reporting (as opposed to the Boing Boing read) that was a relatively insignificant component in a broader pattern of problems.

        I.e. if the incorrect authoring of the worksheet were the only problem, this would not really have affected the outcome of the research. Had everything else been okay, the worksheet error in and of itself wouldn’t have led to grossly incorrect conclusions.

      2. He’s suggesting Reinhart & Rogoff manipulated the data to fit their preconceived notions. The second paragraph refers to how R&R hid the underlying data from public scrutiny.

    1.  Hey, that’s anecdotal, so it’s not really real.  You have to have somebody with proper credentials say something before it’s real.


  2. “Reinhart-Rogoff is the main research cited in favor of cutting public services and spending in bad economic times.”

    On the contrary, the R-R research was hardly ever mentioned — it’s a minor footnote in the reams of research that supports austerity.

    Also, we’ve always been at war with Eastasia.

      1.  Last I checked Paul Ryan was born in Janesville, Wisconsin. Which is in the 1st Congressional District of Wisconsin. That district voted 55% for him as a congressman in 2012. Of course they also voted Obama for President.
        Let’s just make sure we get our facts straight. Besides, there are so many true things you can say about how wrong Paul Ryan is; we don’t need to disguise the truth.

          1. No big deal. I find it interesting that this intellectual cipher couldn’t get his neighbors and fellow townspeople — you know, people who might know him or have interacted with his — to vote for him vs the larger district that might vote based on tribal identity. 

  3. psssst- “Amherst” usually refers to the college, not UMass (at) Amherst.

  4. Looking at the countries cited, I grow more convinced that the military expenditures, even in peacetime (weapons development, logistical expenses for 700+ bases) are a big difference between the US and the countries that live under that shield. Those nations show solid growth even with debt, since the borrowed funds go to pay for things people use, like infrastructure, social services, things that support economic growth as bridges and highways and a good health care and education system will do.

    Guns vs butter and we always choose guns.

      1. You can make butter from guns?

        There’s a basis for the comparators. They’re orthogonal or independent and mutually exclusive of each other. I realize (or hope) your reply was meant in jest but the idea of getting what you want through violence is a litte too topical right now.

        Ike warned us 50+ years ago about this but we have yet to take it in.

        1. “You can make butter from guns?”

          No, but you can make them from soap: John Dillinger is the most famous example.

        2. No, but you can take butter from people who don’t have as many guns.  Not that you’d bother, because the people with the power to make that decision aren’t  short of butter anyway.

    1. Well someone has to show up with the guns at some point in time…and it always seems to be us (usually in places we aren’t wanted, but that’s a topic for a whole different discussion.)

      Of course realistically most of the “bad” guys in the world hate the Western world, which is basically the US…so in a round about way we are the reason we need all these guns…

  5. An interesting point raised elsewhere is that high debt levels and austerity are not opposites.  For example, post WWII Britain had both (high debt and rationing).  And growth.

    So while it may be true that this study doesn’t solidly link debt levels to growth, it has even less to say about links between austerity and growth.  In other words, disproving this study doesn’t automatically prove that austerity is good or bad.  The poof for that lies elsewhere.

    1. The British economy didn’t normalize until 1953 when it finally entered a longer period of growth. The UK had seen GDP crash during the last two years of war, continuing for another two years after the war because of conversion from a wartime economy, then a modest two year bounce back, then recession, one year of growth, then recession again before finally a return to normalcy with five years of growth. Rationing didn’t end until 1954.

      To be frank, the immediate UK post war period is so complicated it simply can’t be distilled into simple narratives like “austerity works”. For one, it was a period of considerable redistribution of wealth, with the founding of the modern welfare state, including the creation of the NHS in 1948. Does that mean “socialism works”?

    2. For example, post WWII Britain had both (high debt and rationing). And growth.

      And Ealing comedies.

  6. Has any country improved their lot after starting austerity measures? We’re a few years after many countries tried that approach, you’d think there would be plenty of examples by now, but I can’t think of any. Theory is nice, but if a measure doesn’t work in practice, maybe it should be dropped.

        1. Barroso, Rehn and the other commissioners know exactly what is
          happening, they just pretend to be surprised when reality diverges from
          official reassurances.

          For Cyprus they officially say it will be no more than a two-year moderate recession before growth resumes in 2015. In an internal paper that got leaked it bluntly states Cyprus will fall into an outright multi-year depression with unemployment comparable to Greece and Spain.

          Obviously they are desperate to save the Euro, which is why they lie through their teeth, but at this stage do they really think it can be saved through austerity? And if they say any other prescription is politically impossible, then how about disbanding the currency union after all? Is it really going to be any more traumatic than this titanic failure?

          1. Ah, but austerity will be reflected by improvements in the stock market, and the stock market is basically the economy, and therefore synonymous with “The Greater Good.”

    1. That’s how we do it in real sciences.  This is a classic example of how a lot of the research you see in social “sciences” is just thinly disguised ideology.

  7. You can’t be serious.

    Seriously, Cory; you’re standing against the serious people. By the Law of what things mean when we use words, you are not serious.

    There’s a right answer and they’ve made it very obvious for you. Begin from a position of total agreement (submission, prostration, fear-peeing) and then we can start to have a serious debate.

    1. The people who cited this are probably the kind of people who begin their screeds with stuff like “It’s simple economics” “economics 101” “You don’t know anything about economics”……..

  8. I think the summary here is misleading.  The errors in this paper were 0.3% or less and updated versions of the paper, with more data, essentially agree with the Amherst version.  A more informed discussion can be found at the Economist:

    “Using the means misses the point of the HAP paper. – It is an argument about the best measure of central tendency. It is basically the old median v mean debate (though in this case it is an average v average-average).The coding error is worth 30bps, and transcription error is worth 10bps – basically nothing. And of course, the new data was not previously available.
    The medians in RR AER 2010 and RRR JEP 2012 are close to the HAP means. Any fair observor would say that all three papers agree — more debt means less growth.

    1. Surely the Economist is a fair observer, having beat the drum for austerity for years.

      The question we’re not supposed to ask is: cui bono?  Who benefits from austerity?  Surely not the poor who become destitute, nor the middle class who become poor.

      I saw a billboard not long ago announcing that “your child’s share” of the national debt is $55,000, or thereabouts.  Horrifying – but false.  Because debt is not parceled out on a per-capita basis, nor does it ever come due.  If it did, my son’s share would be a good deal smaller than Sam Walton’s son’s.

      Since it is the rich who bear the greater burden of taxation, the only ones who benefit from austerity are the rich.  But we aren’t supposed to say it out loud – that would be class warfare.

        1. It’s not The Economist blog claiming it was only skewed by 0.3%. It’s a commenter on the blog named “ricardianambivalence” who wrongly claims that it was only a little bit off. He apparently didn’t wholly read or understand the paper critiquing R&R.

          1. I should have clarified, the quote I gave was from a commentator not the Economist.  However, I do think the “ricardianambivalence” critique is valid, it seems like the errors in the original paper are not all that huge.  He gives a detailed breakdown of the calculations so yes, I think he read the paper in detail.  Instead liberals hear what they want to hear “Oh, those conservatives are all wrong and this paper was way off.”

          2. Basically the “ricardianambivalence” analysis claims that much of the difference comes from the use of mean versus median and doesn’t represent gross errors in the original paper.

          3.  I’ve been itching for an excuse to post that image for a couple days. Don’t bum me out. :)

      1. (Funded) Debt does become due. It’s why the treasury holds regular auctions to carry on funding it.
        $55,000 at the 10 year funding rate needs to pay interest of about $950 per year.
        If funding rates were to return to 2006 levels it would be $2770 a year.
        Not bothered to look up what proportion of the US debt is funded, but if all of the 16 trillion is & debt were to return to 2006 levels, funded via 10 year usts, the annual interest would be 800 billion…

        1. I just don’t get why people don’t understand some of these things…or why the “smart” people don’t explain it in easy to see terms.  It’s like your credit card, which you’ve been carrying a balance on and only paying the minimum month payment.  Except for the US we just keep increasing our limit…*sigh*. Yes somewhere to someone we do owe that money (even if it is ourselves.)

          1. That’s exactly the problem. It’s not like a credit card because you control the money supply. Unlike Greece, Cyprus, Spain, Portugal, Ireland, etc. Similarly there is not a simple household finance model for controlling the money supply. 

      2.  So should we keep increasing borrowing until the entirety of the tax take is spent on servicing that debt? Is then a reasonable time for austerity? The UK is currently spending £43 billion a year (or about 7.5% of the entire tax take) to pay interest on debt. Since 1800 there are 3 big spikes in national debt – all the rest of the time the general trend is downwards. The first was during the Napoleonic wars, the second 2 during the two world wars. Only in the last 15 years has the trend been to rise again.

  9. An Excel spreadsheet loses Belgium, and the resultant warp in data dicks with the world economy? It seems someone in Heaven has hired Douglas Adams as a consultant with a brief to make reality ‘more interesting’.

  10. Carmen Reinhart is a senior fellow at the Peter G. Peterson Institute for International Economics.

    Peter G. Peterson is a billionaire who has promised to spend $1 billion of his money to kill off Social Security and Medicare by convincing people that programs like these will destroy America. Heard of “”? That’s Peterson’s astroturf operation.

    Obviously, this means this was an innocent coding error on Reinhart’s part. Who doesn’t screw up an Excel sheet once in a while, amirite?

    1. Think about all the billions of dollars that have been spent burning the social safety net and electing pro-austerity politicians.  What if we’d spent that money on something useful – say, extending Medicare to cover everyone, or building more good public housing, or something like a modern WPA? 

  11. I once had to do an exhaustive check of a calculation routine for figuring out something about amortizing loans over time.

    The software used was written using Smalltalk and used 11 digits of precision before and 11 digits of precision after the decimal point. It actually did complete iterations over the life of a loan to arrive at precise results.

    There were NO ERRORS possible. If the calculations drifted, the precision was automatically increased. (The routine was responsible for figuring out EXACT payment amounts for multi-billion dollar loans, but converted to Turkish Lira.)

    After a week of trying to see if there were any boundary conditions and running exhaustive checks, I gave up and asked who was reporting the error and how he was getting it.

    It tuned out that the person who reported the error was comparing the results of our calculation routine and the approximations he was getting as a result of running an Excel spreadsheet which used integrals to approximate the results.

    Once I had stopped laughing, I explained to my boss that the Excel integration results were done using Newton’s method of approximating a value and could not be counted for results of any precision. (It might be handy for eliminating repeated calculations but it did so at the prive of introducing unavoidable errors in the results.)

    The work-order for that error ticket was returned to the marketing and sales department with my written request that they never use Excel again.

  12. Even if you don’t make the Excel error, austerity is still something you do during booms, not busts.  All this does is harm R&R’s trustworthiness. 

    In the UK we have austerity explicitly to reduce the deficit and debt. What is happening is we are increasing the deficit and debt. The reason is pretty simple. Osbourne announces austerity and 120 million testicles/ovaries instantly retract amid vows not to spend any money. It’s time to hoard what there is. Literally everyone tries not to spend any money. Business doesn’t invest. Banks don’t lend (as *they* already have debts over 600% of GDP and their best bet is high stakes gambling with computers). Consumers don’t have jobs. Consumers don’t have money. Consumers don’t consume. Falling demand leads to layoffs. Less income. Less taxes paid. Increasing need for former taxpayers to claim social insurance. That’s austerity – a really simple economic death spiral. And the HUGE POINT: government has less income from taxes, more outgoings in benefits claims, and deficit and debt increases at a cost of enormous job losses and hardship for virtually everyone. We are tearing up the fabric of society to *increase* the deficit and debt.

    If government *invested* money – not spent – invested wisely, it could begin a positive life-sustaining economic feedback cycle instead a negative death spiral. We face an existential threat from climate change and there is an energy revolution waiting to be developed. My kids might need a job in 20 years. Get that going. Educate them. Fucking *feed* them. A million kids already go hungry in the UK, and apparently the pain hasn’t even begun!

    The IMF announced it made a mistake with austerity, which Osbourne et al ignored. The IMF miscalculated the multiplier, which describes the amount of economic value resulting from government investment. In the UK, every £1 invested by government would currently return £1.50 of economic value. So the UK is a great government investment. Especially when the government can borrow at historically low low rates, which inflation will later reduce – and government has a lever over inflation, which is also very currently low. It’s a no-brainer. We know austerity raises debt, so at the very least, investing is a more sane option. And there are likely *other* options. But the one thing we do know is austerity now is stupid. Yet Osbourne won’t change course. Most economists in the world agree what he is doing is foolish. I saw Larry Somers struggle to maintain a straight face when asked about it. He literally, nearly burst out laughing.  Osbourne is an Excel error. No feedback. No response. The button is blinking, but the app has crashed.

Comments are closed.