Thursday, July 21, 2011

Shiller: Most People Who Look at Debt to GDP Ratios are Stupid

Yup, that's exactly what Yale economics professor Robert Shiller says in a new paper:

Consider, for example, the debt-to-GDP ratio, much in the news nowadays in Europe and the United States. It is sometimes said, almost in the same breath, that Greece’s debt equals 153% of its annual GDP, and that Greece is insolvent. Couple these statements with recent television footage of Greeks rioting in the street. Now, what does that look like?

Here in the US, it might seem like an image of our future, as public debt comes perilously close to 100% of annual GDP and continues to rise. But maybe this image is just a bit too vivid in our imaginations. 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.

Most people never think about this when they react to the headline debt-to-GDP figure. Can they really be so stupid as to get mixed up by these ratios? Speaking from personal experience, I have to say that they can, because even I, a professional economist, have occasionally had to stop myself from making exactly the same error.
What Shiller is trying to do here is shift a rough view of a country's financial condition and make it appear that many are using it incorrectly as some kind of absolute law .

But let's take a look at this debt-to-gdp ratio and what is does tell us. It's best to start with an example of individuals, their income and debt they may have.

Supposed, I told you that a certain person had $100,000 in personal debt. above mortgage debt. You might think this is high. But if I told you that this debt was owed by Lebron James, who has a multi-million dollar income, the debt wouldn't appear that significant. On the other hand, to a person who has $100,000 in personal debt but only has an annual income of $30,000, things look mighty different.

Thus on an individual basis, a debt-to-annual income ratio can give us a rough picture of the significance of debt. It's the same thing at the national level. In order to get a sense for how significant a country's debt is, it pays to look at that debt relative to the country's income, not because it gives us a hard and fast law about debt and income, but it gives us a rough guide as to how manageable the debt is. Nothing stupid about this at all.

So why is Shiller developing this tortured suggestion that most people are looking at the debt-to-GDP ratio as a belief that a country is bankrupt when it hits a 100% debt-to-GDP ratio? The clue may come further in his paper when he writes:
There is also the issue of reverse causality. Debt-to-GDP ratios tend to increase for countries that are in economic trouble. If this is part of the reason that higher debt-to-GDP ratios correspond to lower economic growth, there is less reason to think that countries should avoid a higher ratio, as Keynesian theory implies that fiscal austerity would undermine, rather than boost, economic performance.
What's going on here is that he is concerned about the debt-to-GDP ratio because he wants governments to take on even more debt, spend more and become and even larger parts of economies. He provides no recommendation at all as to how much further debt a country should accumulate before it stops doing so.

Bottom line: Shiller's argument isn't about people being stupid. It's an attempt by Shiller to hoodwink people into supporting an ever growing big government interventionist state, and cause people to ignore common sense rough guides as to what is going on in the economy.


  1. Wenzel,

    Your comparison seems a little mismatched. In the case of a sovereign and sovereign debt, shouldn't the metric be sovereign debt to sovereign tax revenue?

    The total income of the country (supposedly represented by GDP) only serves an illustrative purpose if the sovereign could essentially tax all that income and use it to service the debt.

    I guess if I were to make a further analogy, it'd be like making a loan to somebody based not on their income, but their extended family's total income, and assuming that if their income isn't enough to service the debt, they'll be able to get their extended family to chip in from their income, as well. This assumption may or may not be valid.

  2. @Taylor Conant

    You are falling into Shiller's trap and making more of the ratio than what it is, a rough guide.

  3. Wenzel,

    I reread what you said and what Shiller said several times and I am not sure how I am "falling into Shiller's trap" or how I am "making more of the ratio than what it is". I fully understand it is a "rough guide". I was specifically critiquing your analogy.

    If a country's debt-to-GDP ratio looks bad, it will look even worse on a debt-to-tax revenue basis because GDP is always greater than tax revenue outside of a fully communist/100% tax regime.

    In fact, I don't even think I understand how your critique is comparable to the point Shiller was making, which seemed to be that using income-per-year is arbitrary because the periodicity (one year) is not special.

    I am not defending Shiller. Now that I think about his remarks, their significance is questionable. If one year isn't a good period to measure something like this, what is? Etc. etc. and further, because GDP is backward looking, Shiller would have to make the case that GDP going forward could be expected to be higher, which would make the ratio look less startling. Instead of doing that, though, he specifically points out that GDP tends to fall in an economic crisis!

    I didn't read the full paper. If you left out some significant part to this discussion then I am speaking in ignorance but if that's the case I am not sure why you would've left out an important piece of Shiller's puzzle.

    I obviously don't get this. I don't think I am falling into any traps because I think whatever metric you use (debt to GDP, debt to tax revenue) as a "rough guide", the picture looks bleak. My point was that assuming a debtor has access to other sources of income than its current or actual sources and then using that as a "rough guide" to figuring out whether or not they can service debt seems to be a confused way to approach the subject as it rests upon an assumption that may not be valid.

  4. @Robert Wenzel

    I do not think Taylor is falling into any trap. I obviously cannot speak for him but I believe he is saying the debt:gdp metric is wrong altogether.

    In your example you compare debt:income. But as Taylor points out (I admit I had never grasped this until just now) gdp does not equal income, unless you are a statist that believes the government owns everything.

    I think I have a better example than Taylor's. Government is a criminal organization whose income is confiscated through taxation and currency debasement. Therefore when using "Crusoe" analogies the individual in question must be considered a thief.

    If we want to examine his debt:income then his income would be the amount he has actually stolen. GDP, however, would represent the amount he could potentially steal.

    Thanks to Taylor it can be shown that Shiller is at least half-right. People who look at debt:gdp are stupid, the correct guide is debt:tax revenue.

  5. Zach Bush,

    I think we're on approximately the same wavelength here. However, I want to be clear that governments can and do tax things other than periodic income. They can also tax accumulated capital/savings.

    I think Bob and I are actually talking past one another hear. Bob is saying "Debt to GDP is a rough guide", and he is of course correct.

    I am saying, "A better guide might be debt to tax revenue (government income vs. "national income")". I think I am right about this, as well, regardless of whether or not I am falling into Shiller's trap, whatever that is supposed to mean.

  6. I also find it amusing that Shiller is saying the debt:gdp is meaningless because of the units but fails to recognize that GDP itself is meaningless because GDP is an array of money prices that cannot be added because they are in different units. You cannot add $2/gallon of milk, $3/gallon of gas, etc.

    This gives further credit to Taylor's assertion that the proper guide is debt:income because debt and income are both expressed in units of money/time.

    As Johnny Drama would say, "victory!"

  7. Actually, I should maybe correct myself here. I said they can and do tax things other than income, but I probably should've said they TRY to tax things other than income.

    I might be remembering my Rothbard incorrectly but I think in Power and Markets he explained that taxes on capital are actually taxes on income (at least in an economy where taxes are payable in money).

    Say you owned some property worth, whatever, $1000, and there was a property tax in place of 5%. Let's say you had no income and no savings. To pay the 5% tax, the only way you'd be able to do that (assuming you couldn't borrow and no one gave you any gift) would be to liquidate your property, as you can't give 5% of its value to the government any other way.

    So, you sell the property and, because it's a really liquid market, you get the full $1000-- this is now an income. So, to pay the 5% property tax you required income, implying this was an income tax in reality, not a property tax.

    What's more sick, now that you generated an income of $1000 you might be taxed on that separately in the next tax reporting period!!

    If Wenzel is still reading this maybe he can chime in and tell me if I am wrong or right about what I just said and whether or not Rothbard said it. I am pretty sure he did but I don't have the time to go look it up right at the moment and Wenzel's mind is like an Austrian rolodex-catalog of information anyway.

    I may have fallen into Shiller's trap but I can confidently say that Bob Wenzel knows his stuff regardless!

  8. @Taylor

    Please feel free to refer to me as Zach.

    And I completely agree that government income can have multiple sources.

    I will summon my inner Hoppe, however, and state that debt:gdp is not merely a rough guide, but is meaningless because GDP can not be calculated (as I demonstrated above).

  9. Oh, and let's not forget the bogus "government spending" component of GDP.

    So dissipating (sovereign) capital is considered wealth production?? So the more they spend, the better their debt ratio looks?? Even though it increases the debt?? (I know there is a diminishing return there, but it does serve to diminish the ***APPARENT*** harm).

    I guess you've gotta be a Keynesian genius to get THAT one.

    I'm just a lowly tax payer. :(

  10. @Taylor Conant and Zach Bush

    Shiller is attacking the debt to GDP ratio as a guide becasue it is not an exact measure and it scares people when it is around 100%---as it should. Overall, as a rough guide, it is very helpful in signalling to me that a country with a 153% debt to GDP ratio is likely in serious financial trouble.

    Shiller doesn't want to use this measure. I'm sure he would be more than happy to go along with you guys and look at tax revenues, and then determine how much they should be raised to justify even more debt for a country.

    You are going down his road.He wants the micro-managers to come in and look at tax revenue flow. Right where you guys are going. He is against the debt-to-GDP ratio becasue it is only a rough guide, which really can't tell you enough about how much debt a country can sustain.

    Shiller's attack is strategic. It's much easier to increase government debt bringing in the mechanics to justify borrowing on revenue formulas, than it is when people have a gut feeling that a 153% debt to GDP ratio is high.

    As libertarians are seriously proposing that we look at government tax revnues to see how much debt should be justified?

    If you are doing a financial analysis to understand a govrnments debt burden that is fine.

    But, if you are fighting expansionary government, why would you want to bring the tax revenue number at all? It's either going to result in either more debt immediately or a call for higher taxes so that more debt can be raised. This is the entire purpose of Shillers paper!

  11. Robert Shiller is a political statist terrorist trying to convince the mentally retarded public school tax cattle that they are stupid and to be good little obedient slaves and look away. The fact that the terrorist state is borrowing is bad enough to make thinking people want to elliminate the statist terrorists before they murder all of us.

  12. @Robert Wenzel

    I do not see how you conclude that by us saying that the proper figure is debt:tax revenue that we are saying that tax revenue should be increased to justify more government debt. That is a pretty bold leap.

    The fact that GDP does not equal income and tax revenue does (and also that GDP is meaningless) are truths. To reject this is to reject reality.

    Whether or not one believes that tax revenue should be increased is a value judgement and an entirely different discussion.

    In my above example I stated that goverment must be modeled as as a thief, I think that would be enough to draw the distinction between myself and Shiller. To be more clear, as one who favors the free-market (a voluntarist, anarcho-capitalist,whatever), I want the government debt:income ratio to be a null set because in the free-market there can be no taxes.

    Now it should be quite clear that I do not agree with Shiller because I wish to expand government. I agree with with him only in so far as debt:gdp is meaningless, but not because debt:gdp is meaningless but because GDP itself is meaningless.

    Shiller is actually falling into the Misean-Rothbardian trap that index numbers are meaningless. If he extended his unit analysis further he would arrive at the conclusion that all Keynesian aggregates are invalid and therefore so is Keynesianism (and monetarism) itself.

    By suggesting GDP is a useful guide you have fallen into the Keynesian trap of using aggregates that do not exist.

  13. Wenzel,

    As libertarians are seriously proposing that we look at government tax revnues to see how much debt should be justified?

    I am appalled and confused that you should even feel the need to ask, and that you seem to have misinterpreted intent from the get go. Are you not feeling well today?

    To answer your question, I am looking at it from a "what is logical" sense, which you call "financial analysis". My whole point was that in terms of financial analysis your metaphor was not accurate.

    Notice in my original comment I didn't make any mention of Shiller or his trap. I was commenting solely on the accuracy of your metaphor. I agree with you that Shiller is attempting a trap. That doesn't mean I am incorrect in asserting that your metaphor was inaccurate in terms of financial analysis.

  14. @Zach Bush

    First, Shiller no where, no how implies that GDP is meaningless. He says the debt to GDP ratio is.

    As a rough measure debt-to-gdp is fine, that is why Shiller wants to kill the notion. It prevents him from being able to call for more debt.

    That was the purpose of my post, responding to Shiller's attempt to kill the value of debt-to-gdp as a measure that is justifiably scaring the hell out of people.

    My point was not intended to go beyond that and create more efficent ways to measure stress in debt levels---which Shiller would use to justify increasing debt levels.

    This is not denying reality. If you would have become aware of the Manhattan Project and further understood a more efficent way to deliver an atomic bomb, would you have been denying reality if you did not discuss with the bomb builders what you understood?

    @Taylor Conant

    I repeat, Shiller is trying to kill off the notion that the debt-to-GDP measure has value, that was the basis of the post. I only later added in a comment that I understood a financial evaluation could take a more in depth view when you brought it tax revenues in your post, but it remains a dangerous side issue relative to the main point--which I think Shiller would welcome.

    I am feeling fine today, thank you for asking.

  15. Wenzel,

    I am feeling fine today, thank you for asking.

    Good, just checking ;)

    We can't afford to have you taking any sick days

  16. And if you are wondering which metric people would find scarier...

    Debt:Tax Rev vs Debt:GDP

    Which figure makes the government look more insolvent?

    Disclaimer: Using GDP is for demonstration only. I still maintain that GDP is a meaningless number.

  17. Zach Bush,

    Maybe I can't read the graph well but wouldn't Federal expenditures - Federal receipts = Federal deficit, not federal debt?

    because if it were debt, the ratio should be a lot higher than 60%

    (maybe I am reading it wrong)

  18. @Zach Bush

    This is my point, the chart then brings into direct discussion whether revenues need to be raised. I'm sure Shiller would find your chart a lot more useful than the debt-to-gdp chart. Robert Reich and the President would for sure.

  19. Wenzel,

    It should be useful to us, too, in illustrating just how much of the productive capacity of the country the federal government consumes (and thereby wastes), making us all poorer in the process.

  20. @Robert Wenzel

    Please feel free to call me Zach.

    The "denying reality" was in response to your assertion that Tyler and I were "falling into a trap" by showing that income does not equal GDP. In your original example you used the debt and income of one person then immediately switch to debt and gdp for a national government. But as Tyler originally commented this is not the same comparison. If you want to compare an individuals debt:income to a national governments debt:income then you must use debt:revenue to remain consistent. Did you not just the other week slam Keynes, correctly, for his inconsistent use of the word savings?

    If you want to compare debt:gdp of a government and individual you would have to compare federal debt:gdp vs an individuals debt:gdp.

    What the chart shows is govt debt:income. I believe it is beneficial because it shows the reality of the situation: to decrease the ratio you can either raise tax revenue or cut spending.

    It also shows the true severity of the situation. This is the Toyota Way of problem solving, addressing the root cause. To continue masking the problem will never solve it.

    If statist pigs like Shiller and Reich wish to use this data to convince people to raise taxes then so be it. You and I can use the same data to advocate reductions in spending. The outcome is not determined by the data, it is determined by whose ideas (voluntarists vs involuntarists) are accepted by US citizens.

  21. Premises:

    *The federal government produces nothing (the combination of all services it provides runs at a net loss)
    *The federal government never truly cuts spending
    *To alleviate debt the government must tax more, sell assets, or seize assets.
    *The US government is not composed of looters, so it will raise taxes or sell its assets.

    The main point by Shiller is (I disagree):
    Debt to GDP ratio is meaningless because GDP is production based on units time and debt can be frozen, so the debt can always be eliminated with some government actions.

    The trap: To pay off the debt at any given time, based on our premises, the government must raise taxes.

    Therefore the figure is not meaningless, and in fact is an excellent indicator of the health of a nation. If a debt to GDP ratio is 5000%, and a crisis occurs such that the status quo is majorly disrupted, then even if the government taxed the economy such that the entire GDP was absorbed, the government would not be able to eliminate its debt, and the nation would descend into anarchy, since all sectors would be destroyed.

  22. You guys are wrapped around the axle.

    I made my millions by shorting the U.S. stock market 17 months before the fall '08 crash. How did I see it coming? I saw graphs of household debt/GDP over time, and that the ratio today was much greater than what it was in '29. I then looked at GDP and household debt in the '30s, and saw that both moved down.

    I concluded that wherewithal to repay debt peaked in '29, then collapsed in the '30s. Such caused a depression, and a big, then sustained, drop in the stock market.

    We are having a replay of such, now.

    Debt/GDP works well as an indicator of a problematic state of economic affairs.

  23. With all due respect I believe it is far more beneficial to statists for Austrian economists to continue to use GDP in econ/financial analysis because it gives credit to a meaningless number that is the foundation of keynesian and monetarist economics.

    GDP is defined as the the total dollar value of all goods and services produced in a given region over a certain time period. One of the first things taught in mathematics is that in order to add/subtract two quantities they must be homogenous (ie. have the same units). The dollar value of a good, x, has the units of $/x. The dollar value of a good, y, has units of $/y.

    It is clear that it is impossible to add the dollar value of the different goods x and y. Macroecon then performs a subtle misdirection. To calculate GDP the money incomes (or money expenditures) are used. Ignoring all other problems with the income and expenditure GDP models, the problem with this is that money income and dollar value of a good are two completely different things.

    Dollar value is a price, the rate at which money is exchanged for a good. Money income is just the money unit itself and has no relation whatsoever to the value of the goods in existence. GDP then is only a certain method to count the money supply. As any good Austrian knows however, any amount of money can support an economy and does not measure wealth.

    Exposing GDP for what it really is explodes many Keynesian and Monetarist myths:

    1) an economy can only grow if the money supply expands

    2) index numbers can be used to manage the economy

    3) government spendings expands the economy

    I am sure there are more but I am typing on my phone and my hands are starting to hurt. :)

  24. @anonymouses

    You are right. My choice of the use of the word meaningless was misguided. Meaning is a subjective value and by claiming GDP to be meaningless I inserted my own value judgement.

    I should have said GDP, defined as the $ value of all goods and services in a region, is misleading because:

    1) value is subjective and can not be measured

    2) the money prices can not be added because they are not homogeneous

    3) the method used to calculate GDP are money incomes and are not the same as money prices

    It follows that GDP is just a money supply figure and is not a measurement of wealth and can not be used as indicator of economic growth. It only shows how money has been allocated over a certain time period.

    Thank you for the comments as they helped me refine my argument.