Is the eurozone’s problem excessively small government?

Commenter Patrick pointed me to new book on the secular stagnation problem.  Lots of contributions by well known Keynesians. A chapter by Guntram Wolff discusses the causes and cures for the eurozone stagnation.  Here’s the abstract:

The persistence of low Eurozone inflation undermines private and public debt sustainability – especially in the periphery where the overhang is greatest. However, since bubbles and unsustainable borrowing supported demand before the Global Crisis, this chapter argues that higher inflation cannot be a permanent cure for secular stagnation. Instead, a targeted quantitative easing programme and increased public investment would help rebalance Eurozone demand. At the global level, population growth in Asia and Africa will provide ample investment opportunities if they can be fully integrated into the world economy.

In the chapter he repeats the often heard claim that relying solely on monetary policy will prop up aggregate demand at the cost of creating serial bubbles.  That tells me that belief in bubbles cannot be viewed as a harmless quirk, like belief in Santa Claus, but rather is something that can seriously damage policymaking.  I don’t believe that bubbles exist.  But even if I am wrong, there is absolutely no evidence that monetary policy is responsible for asset price bubbles, which are actually less common during periods where monetary policy is very easy (like the 1960s through the early 1980s.)  Most people confuse low interest rates with easy money.

I also found it odd that someone would claim that more government spending (or more specifically an increase in public investment) is the answer to the eurozone’s stagnation.  Here are some of the developed countries with the lowest levels of government spending:

Singapore (17.1% of GDP)

Hong Kong (18.5%)

Taiwan  (22.6%)

Switzerland (33.8%)

Australia  (35.3%)

I don’t know the eurozone ratio, but Germany seems well below average at 45.4%, most countries are close to 50%, and France is on the high end at 56.1%.

In fairness, the abstract mentions population growth at the end, which is high in Australia and Singapore.  But France has a relatively high birth rate for a developed country (much higher than Germany) and immigration also adds to its population growth.  And yet France is clearly stagnating.

I do buy into Wolff’s claim that the eurozone is in the midst of a major stagnation—indeed even worse than the US.  However I very much doubt whether this has anything to do with a lack of government spending or even government investment.  AFAIK, there is no “Great Stagnation” in any of the low government spending developed countries (which also tend to run budget surpluses, or small deficits).  My hunch is that while demand has recently depressed eurozone output, there is a longer term supply-side problem in the eurozone that is more likely related to high government spending, combined (in some cases) with an inefficient public sector.

A few notes on the inevitable counterexamples:

1.  Yes, many poor countries have low levels of government spending (but not Brazil!)  Their economies are not developed enough to easily extract large amounts of taxes from poor peasants. There are large informal sectors, where people work off the books.  Nonetheless, significant wealth creators often do face high implicit MTRs in these countries.  My point is that at the top end, just looking at developed countries, the small government countries have done better.

2.  There are some big government economies that are doing alright, such as Germany and the Nordics.  That shows that size of government is not the only thing that matters.  But Germany would do even better with a Swiss-style economy.  Also keep in mind that very few countries have the low levels of corruption seen in the Nordic countries.  In southern Europe, eastern Europe, and most of the rest of the world, any attempt to push government spending to very high levels will get immersed in corruption, and the resulting waste will be a drag on growth.


Tags:

 
 
 

30 Responses to “Is the eurozone’s problem excessively small government?”

  1. Gravatar of Maurizio Maurizio
    19. August 2014 at 07:39

    About your claim that asset bubbles might not exist, and that there is no evidence that monetary policy is responsible for asset price bubbles, I have a question.

    Suppose (for the sake of argument) that, in the next few months, US stocks were to crash, and they required another round of QE, bigger than the previous one, to get back on track; or more in general, suppose that we discover that US stocks are unable to keep the current levels, without a succession of QEs of ever increasing size. Would this prove, in your view, that 1) stock bubbles can exist, and 2) that the QEs have created one?

  2. Gravatar of brendan brendan
    19. August 2014 at 07:40

    Stock valuation peaks: 1929, 1966, 2000
    Stock valuation troughs: 1933, 1974, 1982, 2009

    The volatility of NGDP (trailing, current and expected) is the dominant determinant of aggregate valuations. No other factor explains much variation; the level of rates, or the level of trend RGDP growth don’t much matter. Common sense confirms- willingness to hold stocks depends on recency of last 401K implosion.

    If macro scale bubbles exist they’re caused by long stretches of superb monetary policy; inverse bubbles (busts?) by long stretches of incompetence. This is obvious to anyone who hasn’t had the common sense and historical curiosity trained out of them by an econ PhD.

    If micro scale bubbles exist (they do- but then, who cares from a macro perspective) they’re caused primarily by the interaction of short sale constraints and sexy stories/sectors/techs. (Wide belief dispersion w/ pessimist info silenced.)

    In terms of absurdity and potential cost, the idea of Yellen acting like George Soros is a friggin gem.

  3. Gravatar of Ironman Ironman
    19. August 2014 at 08:03

    ssumner writes:

    But even if I am wrong, there is absolutely no evidence that monetary policy is responsible for asset price bubbles

    It depends upon what you mean by asset price bubble. By our definition, a bubble can be said to exist whenever the price of an asset that may be freely exchanged in a well-established market first soars then plummets over a sustained period of time at rates that are decoupled from the rate of growth of the income that might be realized from owning or holding the asset.

    So to tell if there is a bubble in the stock market, for example, you would examine the relationship between stock prices and dividends per share for such a disruption. You should then be able to do an event analysis to identify when the disruptive event began and when it ended, from which you should be able to identify causal factors.

    When you do that with the Dot-Com Bubble in the U.S. stock market, which fully ran from April 1997 to June 2003, you will find that changes in U.S. tax policy are responsible for the inflation of the bubble and the restoration of stability in the market – not monetary policy. Those changes altered the returns for certain types of investments with respect to others, which then led to the positive feedback cycle that resulted in what proved to be the unsustainable inflation of the Dot Com Bubble.

    That’s not to say that monetary policy could not initiate something that looks like a bubble. As we’ve seen with QE and when it’s been turned on and turned off in the past, it can most certainly have an impact upon stock prices that very much looks like a bubble. And if not, the Fed certainly wasted a lot of effort in targeting stock prices to measure the effectiveness of their QE policies.

  4. Gravatar of ssumner ssumner
    19. August 2014 at 08:28

    Maurizio, No, it would depend why stocks crashed. Suppose they crashed because NGDP growth expectations plunged.

    Brendan, Can you translate that comment into English?

  5. Gravatar of Luis Pedro Coelho Luis Pedro Coelho
    19. August 2014 at 09:18

    In Europe, Sweden is not a big outlier wrt government spending
    (http://en.wikipedia.org/wiki/Government_spending#As_a_percentage_of_GDP).

    Sweden 51.2%
    Portugal 49.4%
    Italy 49.8%
    Greece 51.9%
    Spain 45.2%

    Given recent trends (gov shrinking in Sweden, gov growing in Portugal), I expect Portugal to overtake Sweden by this year or next.

    (And the supply side is much more free market in Sweden).

  6. Gravatar of MB. MB.
    19. August 2014 at 09:40

    If stock markets crash because expectations plunged, then isn’t it a bubble? Or a sunspot or whatever the name you give to fluctuations non-driven by changes in fundamentals?

    Bubbles is a useful theoretical concept, even if it might be difficult (impossible?) to really forecast such phenomena.

  7. Gravatar of TallDave TallDave
    19. August 2014 at 10:06

    Anytime anyone acting in an official capacity in a modern economy suggests “demand” can be boosted with public spending, eyes should roll and heads should follow. This is not the 1930s.

    I think bubbles can only exist in the sense that expectations often change, but then every positive price fluctuation could be described as a bubble so the term really adds little meaning.

  8. Gravatar of Maurizio Maurizio
    19. August 2014 at 10:19

    Maurizio, No, it would depend why stocks crashed. Suppose they crashed because NGDP growth expectations plunged.

    Thanks. I’m a bit confused. Is this like saying “suppose stocks crashed because the market did not believe QE4 would take place?”

  9. Gravatar of brendan brendan
    19. August 2014 at 10:49

    Yeah, sorry, I’ll try again.

    Forget bubbles, and let’s just talk about stock valuations. Look at historical valuation peaks and troughs. What caused them? What explains variation in p/e’s?

    The most important factor, by far, is the amount of preceding (trailing 1-10 years) NGDP volatility; the relation is inverse, of course.

    Low P/E’s, with out exception, are caused by periods of exceptionally volatile NGDP; high p/e’s by periods of low NGDP volatility.

    Stock valuation peaks: 1929, 1966, 2000
    Stock valuation troughs: 1933, 1974, 1982, 2009

    Three generalizations that fit the historical record:

    1) When – and as long as- NGDP growth is stable, valuations tend to rise.
    2) Periodic NGDP instability whacks valuations; when shocks bunch together in time, as in the 70’s, valuations get very low. But in the absence of NGDP shocks, valuations tend to rise w/ out limit; long periods of calm produce “bubble” like p/e’s (late 1990’s).
    3) In a stable NGDP environment, the rate of growth of valuations is inverse to valuations. If p/e’s are lower, valuations rise rapidly; if p/e’s are high, they rise more slowly.

    Bland, yes. But there are few other generalizations that fit the record. The idea that low rates cause high P/E’s doesn’t fit. RGDP rates can’t be that important, since we’ve seen abnormally high P/E’s in the last 30 years.

    This makes sense if you observe people and you introspect. “What are the odds I lose 50% of my money” is what people wanna know, and their guess is based on what they can easily recollect- what’s happened in the last 10 years. Survey data confirms this kind of recency bias.

    Anyway, the point is that, if bubbles exist, they’re caused by periods of superb monetary policy. For central banks to reduce valuations they will have to cause NGDP instability- which is nuts.

  10. Gravatar of brendan brendan
    19. August 2014 at 11:05

    Or, more concisely, and translated into expectations lingo, valuations are determined by a) expected NGDP growth and b) the confidence interval around expected NGDP growth.

    One way of thinking about why the Evans Rule boosted equities with out boosting the rate of NGDP growth much, is that the confidence interval around expected NGDP growth narrowed.

    OK, I’m verging on sophistry. I’m done.

  11. Gravatar of TravisV TravisV
    19. August 2014 at 11:18

    “Inflation Rises At Slowest Pace Since February”

    http://www.businessinsider.com/july-inflation-report-august-19-2014-8

  12. Gravatar of Kevin Erdmann Kevin Erdmann
    19. August 2014 at 12:37

    US govt. spending per capita is similar to northern Europe. It is only lower as a % of GDP because our per capita GDP is higher. Imagine what a boon it would be if the US govt provided the same level of services as the northern European countries do with that spending.

  13. Gravatar of benjamin cole benjamin cole
    19. August 2014 at 15:53

    Excellent blogging.
    Why is it free marketeers contend that private-sector decision-makers lose their marbles when interest rates are nominally low?
    As tight money leads to low nominal interest rates, that is quite a conundrum…

  14. Gravatar of ssumner ssumner
    19. August 2014 at 18:30

    Ironman, I use Fama’s definition–predictable markets.

    MB, You said:

    “If stock markets crash because expectations plunged, then isn’t it a bubble? Or a sunspot or whatever the name you give to fluctuations non-driven by changes in fundamentals?
    Bubbles is a useful theoretical concept, even if it might be difficult (impossible?) to really forecast such phenomena.”

    It’s not a useful concept unless you can predict markets with it. See Fama’s new AER piece. And no, expectations are a fundamental factor if they are rational.

    Brendan, Yes, it makes sense that valuations would be high when investors expect more stability in NGDP growth.

    Kevin, Good point.

    Ben, Yes, their analysis leads to a dead end.

  15. Gravatar of maxk maxk
    19. August 2014 at 20:16

    How fundamentally meaningful are government spending numbers as a percentage of GDP? For example, suppose country A funds everyone’s retirement via a tax and distribution system. Everyone is taxed during his working years, then receives checks during retirement. The dollars each worker gets are scaled by contributions, but with mild redistribution. Low wage workers get a bit more than they paid in; high wage workers get a bit less; average wage workers get back about what they put in. Conversely, country B has a requirement that everyone set aside a fixed percentage of wages into a private retirement account. In addition, the highest wage workers pay a small tax that is used to supplement the retirement accounts of the poor. Taxes for retirement in country B are much lower than in country A, but the ultimate flow of dollars to individuals is quite similar.

    Or imagine a similar arrangement with medical taxes vs. individual medical accounts.

    Sure, you can prefer the style of B over A for other important reasons, but the implication of the large quoted differences in government spending by gdp is that some countries are somehow doing the same things for much less money. When in fact in the ‘low government’ countries, some important functions are just being done elsewhere, perhaps covered by roughly equal private dollars that are still under considerable government control.

    Or else they’re done by purely private dollars, not under government control, with benefits from that (individual watches his dollars more closely) but also losses (more variance in individual outcome, less insurance, less protection against fraud).

    It’s just hard for me to imagine that the quoted numbers are in any real sense comparable. The devil is lost in the details.

  16. Gravatar of TravisV TravisV
    19. August 2014 at 20:44

    Prof. Sumner,

    Have you seen that Brad DeLong wrote this?

    “All Republican macroeconomists who want to preserve their reputations and continue to be of note should be strongly behind the market monetarists–should have the backs of James Pethokoukis, Scott Sumner, Ben Bernanke, David Beckworth, Michael Darda, Lars Christenson, Ramesh Ponnuru, and company. Hell, on the question of what Japan should do when its liquidity trap started, Milton Friedman had their back.

    But they don’t.

    It is not good for Bob Lucas to get confused and think that the marginal propensity to consume out of extra government debt this year is minus one. It is not good for John Cochrane to claim that since future government primary surpluses are fixed and real interest rates rapidly revert to normal that the price level moves proportionately with the stock of nominal government debt. It is not good for Ed Prescott to claim that the Great Depression was an equilibrium response to Herbert Hoover’s anti-market regulatory and tax policies. It is not good for John Taylor or Marty Feldstein to be overestimating the dangers of inflation for seven years in a row.

    Is it?

    So when Krugman writes:

    The market monetarists have a hopeless task. James Pethokoukis… needs to ask why that obsession persists… after five years of utter empirical failure…

    I think Paul misses the point: Academics of note and reputation do not become such without strong desires to maintain and enhance their reputations and become of greater note by publicly getting it right and getting it right in ways that potentially influence policy. When we think those who seek to be our political masters are getting it wrong, those of us who find ourselves on the Democratic side of the aisle have no problem saying so. Even though we know it is a hopeless Charge of the Light Brigade, we have no problem horsing up. And we have little problem trying to mark our beliefs to market when we get it wrong.

    The way the world is supposed to work is that politicians are supposed to be the slaves of some defunct economist. Economists are not supposed to be the slaves of some underbriefed politician.”

    http://delong.typepad.com/sdj/2014/08/monday-delong-smackdown-the-wellsprings-of-bad-monetary-economics-in-goldbugism.html

  17. Gravatar of Daniel Daniel
    19. August 2014 at 23:03

    Suppose (for the sake of argument) that, in the next few months, US stocks were to crash, and they required another round of QE

    If QE is required, that means a monetary contraction occured.

    A fall in asset prices is exactly what the sticky wages model would predict in such a case.

    How exactly does that prove the existence of “bubbles” ?

    In fact, what is a “bubble” ? Can any of you offer me a non-circular definition ?

  18. Gravatar of Maurizio Maurizio
    20. August 2014 at 00:21

    If QE is required, that means a monetary contraction occured.

    You mean that the monetary base must have contracted? I don’t see why. You mean stocks cannot crash if the monetary base just stays constant, i.e neither shrinks nor grows?

    In fact, what is a “bubble” ? Can any of you offer me a non-circular definition?

    This does not immediately look circular to me: some stocks are said to be in a bubble if they can’t keep the current market value unless new money keeps being printed (and spent on them).

    i.e., if they crash as soon as the money printing stops.

    (if everything else stays constant, of course. If people shift preferences and start spending relatively more on those stocks, of course they can keep their value, but this is beside the point. We have to assume the money stops increasing and everything else stays the same.).

    Does this look circular?

  19. Gravatar of Daniel Daniel
    20. August 2014 at 00:40

    You mean that the monetary base must have contracted?

    Last time I checked, it was M * V

    If M1 = 3 * M0 and V1 = V0/4, then M1*V1 = 3/4 * M0*V0

    THAT is a “monetary contraction”.

    You mean stocks cannot crash if the monetary base just stays constant ?

    I’m saying you need to understand the difference between “supply-side” issues and “demand-side” issues. Which you refuse to do.

    they crash as soon as the money printing stops.

    Like I said before – THAT is exactly what the sticky wages model would predict.

    After 25 years of 5% nominal growth, don’t you think a sudden switch to 0% growth would cause problems ? Wtf, is “stickiness” so hard to grasp ?

    Try harder. This time with a bit of critical thinking.

  20. Gravatar of Maurizio Maurizio
    20. August 2014 at 01:05

    Thank you Daniel. That helped.

  21. Gravatar of ThomasH ThomasH
    20. August 2014 at 03:45

    I have not been able to understand if the “Keynesian” (Summers-Krugman) type of stagnation is based on the assumption that faced with the ZLB for short term interest rates monetary authorities are unwilling to provide monetary stimulus in the appropriate amounts by purchasing something other than short term government paper or on the assumption that governments, faced with low interest rates will not invest in projects with positive NPV’s? Given the experience of the US since 2008 and even more the experience of the Euro zone, both of these assumptions about policy are not unreasonable, but do these models give “secular stagnation” under different policy regimes?

  22. Gravatar of ThomasH ThomasH
    20. August 2014 at 04:15

    @ Tall Dave
    “Anytime anyone acting in an official capacity in a modern economy suggests “demand” can be boosted with public spending, eyes should roll and heads should follow. This is not the 1930s.”

    I think this is the wrong way to look at it. If during a recession interest rates on long term government bonds fall to near zero, many projects that do not have positive NPV’s during “normal” times will have at the lower discount rate. If governments act on good neoclassical microeconomics principles they will increase investment, a component of final demand, and it will look like “fiscal policy.”

  23. Gravatar of ssumner ssumner
    20. August 2014 at 04:43

    maxk, That’s a good argument, and it largely explains why the Nordics do so well. Most of their “big government” is the more efficient type. Denmark comes in around 10 on the Heritage ranking of economic freedom, despite high government spending. That also explains why Singapore is so low (partly, not entirely.)

    Maurizio, Think of QE affecting stocks only to the extent that it affects the future expected path of NGDP. And that is a fundamental factor.

  24. Gravatar of J Mann J Mann
    20. August 2014 at 05:28

    Daniel asks:

    “In fact, what is a “bubble” ? Can any of you offer me a non-circular definition?”

    Here’s my understanding of what most people mean when they use the term.

    1) Dutch tulips: My understanding of what people often mean is when a price rises substantially above its underlying value solely because people are hoping to sell to the next person.

    – So if someone discovers a cheap home metho of manufacturing diamonds tomorrow and the price of diamonds crashes, that’s not a bubble, that’s new information.

    – And if the price of beanie babies goes up because collectors want them, that’s not a bubble.

    – And if the price of tech stocks skyrockets because people honestly think that one stock in ten will turn out to have a gigantic future income stream, then that’s not a bubble either.

    – People’s impression of a bubble is when all the informed investors “know” that there’s no consumption use or income stream for something that can justify its price, but they keep running the price up because they think that other people will continue to buy. Those other people can either be dumb money or smart money planning to get out before the crash.

    I’m sceptical whether this happens often enough to be interesting.

    2) Unreasonable price increases, followed by crash. More colloquially, “bubble” is used to refer to a sustained increase in price, followed by a crash. (E.g., tech bubble, housing bubble, etc). This certainly happens, but isn’t that interesting.

  25. Gravatar of Daniel Daniel
    20. August 2014 at 05:49

    J Mann,

    I know what people think of when they speak of “bubbles”, what I’m saying is that none of those definitions is of any practical policy use.

  26. Gravatar of J Mann J Mann
    20. August 2014 at 06:57

    “Useful,” huh? That’s raising the stakes from “non-circular.”

    I don’t personally believe that there are any useful definitions of bubbles, but as I understand the argument:

    (1) If case 1 “tulip bubbles” are identifiable when you are in them, you should be able to make money. You’d have to structure the play to withstand a potentially long run of bubbledom, but if it’s truly a bubble, you should be able to do it.

    (2) Sumner and Obama think that appropriate central banking can reduce the frequency of case 2 bubbles.

  27. Gravatar of TallDave TallDave
    20. August 2014 at 12:03

    ThomasH

    Well, true counter-cyclic policy rarely happens because governments are far more likely to over-invest during booms. But I think there are some problems with that theory — if NPVs are actually better during recessions, why don’t private investors see that too and leap into action to invest their money? And the answer is because they see less possibility for growth than they did before the recession. And that same argument applies to government investment — the bridge may be cheaper when people expect less growth, but its projected usefulness also falls.

  28. Gravatar of TallDave TallDave
    20. August 2014 at 12:07

    J Mann,

    Unreasonable price increases

    No such thing. Is it reasonable to pay LeBron ~$100M/year because of his skill at putting a ball in a hoop on a hardwood court while following a bevy of arbitrary rules? Why are old dollops of paint on canvas reasonably worth ~$100M because they were put there by Monet? And the answer is always “if enough people think so.” And people can change their minds, too.

  29. Gravatar of TallDave TallDave
    20. August 2014 at 12:14

    underlying value

    Same problem. For instance, diamonds’ value stem primarily from people’s appreciation of sparkly things. If people stopped liking sparkly things, they would have little value. All utility lies in the eye of the beholder!

  30. Gravatar of Daniel Daniel
    20. August 2014 at 13:23

    J Mann, I don’t know about you, but way back in high-school logic class I was taught that a valid definition must meet three criteria

    1) genus
    2) differentia
    3) non-circularity

    So far, the people who talk of bubbles either use outright circular logic – “a bubble is a bubble, I’ll know it when I see it” – or, when pressed for specifics, say stuff like “an asset whose price goes down when NGDP contracts” – which is kinda like saying “everything is a bubble”.

Leave a Reply