Monday, September 13, 2010

Can the Nobel Prize be Revoked?

The answer to the question is no (see here.). Alfred Nobel originally set up five Nobel prizes: Physics, Chemistry, Medicine, Literature, and the Peace Prize. How Literature fit in, it's not clear (maybe Alfred liked to read by the fire in the evenings), and I'm not sure what stories lie behind the omission of Mathematics and Biology, for example. However, at some point, the Swedish Central Bank saw fit to fund the Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel, often referred to as the Nobel Prize in Economics. The natural scientists, writers, and peacemakers may not feel comfortable with us economists, but there we are.

Now, in 2008, Paul Krugman was granted the Nobel Prize in Economics. In giving the prize, the Nobel Prize committee always cites some specific work or body of work that justifies what they are doing. In Krugman's case, he got it "for his analysis of trade patterns and location of economic activity." Krugman is a recognized expert in the field of international trade, and the Committee, after input from members of the economics profession and deliberation, decided to award Krugman the prize.

The Nobel Prize recognizes important and influential contributions to our knowledge. To pass the bar for the Nobel prize, we have to think of these contributions as truly pathbreaking. The prize-winner must somehow have made us all better off in a significant way. I don't think that anyone imagines that conferring the prize on an individual will make that individual work harder for humanity. What we are thinking is that young economists might see the recognition of a Nobel Prize (and the million bucks too) at the end of the rainbow. In spite of the fact that the chances of getting a Nobel are tiny, perhaps young economists at the top of the profession will all work just a little harder, and the benefits for society will be worth what the Swedish Central Bank spends on the thing every year.

Now, I asked a question in my title, and I mentioned Krugman, so you know what is coming now. We might forgive Paul for his lapses in macroeconomics. He's a trade guy after all, and we don't expect him to be up to speed in modern macro. However, the prize he got was for his work in international trade, and if he is writing about issues in that field, we expect a lot. That's what makes today's NYT column so disturbing. Part of what we do in undergraduate economics classes is to dispel myths, and to show how some basic tools from economic science can shatter ideas that can seem intuitively plausible. Krugman's column takes economic knowledge backward. This is not an attempt to enlighten lay readers by showing them how some simple economics works. It's an exercise in bad economics put forward as mainstream thought, and beyond reproach. If it were in my power, I would love to take the guy's prize away. Of course I can't, and unfortunately no one else can, either.

First, Krugman says this:
China is deliberately keeping its currency artificially weak.

The consequences of this policy are also stark and simple: in effect, China is taxing imports while subsidizing exports, feeding a huge trade surplus.
First, of course there is nothing "artificial" about China's exchange rate policy. The Chinese government and its central bank are free to intervene as they see fit in financial markets, and those actions are going to have an effect on the exchange rate of the renminbi against the US dollar. Currently, though the Chinese claim that the renminbi/US dollar exchange rate is flexible and market determined, they appear to be essentially pegging it. But so what? A basic principle of international economics is that a country cannot "subsidize" its exports through what are essentially monetary policies determining its nominal exchange rate. The prices of goods sold by China on world markets are determined in world markets in US dollars. There is nothing that Chinese exchange rate policy can do to affect those prices.

As to the claim that China's nominal exchange rate policy is "feeding a huge trade surplus," the state of China's current account balance has little to do with exchange rate policy. In China, financial intermediation is very inefficient. Banking in China is dominated by a few large state-owned banks, and those banks are not very good at channeling domestic savings into high-return domestic investment projects. The savings of Chinese households is huge. What happens to that savings? Much of it leaves the country in the form of US Treasury securities and other foreign assets, resulting in lower real interest rates in the rest of the world (than would otherwise exist), as compared to the high domestic real interest rates in China. Some of the savings that flows out of China comes back in as foreign direct investment, but there are barriers to that flow. The result is a current account surplus in China - Chinese national savings exceeds investment. The primary cause has nothing to do with the nominal exchange rate. The Chinese current account surplus would be much smaller if domestic financial intermediation in China were much more efficient - and our interest rates would be higher.

Now, Krugman perhaps is anticipating this criticism:
You may see claims that China’s trade surplus has nothing to do with its currency policy; if so, that would be a first in world economic history. An undervalued currency always promotes trade surpluses, and China is no different.
The big question here is what "undervalued" means. Relative to what? If I can decide what undervalued is, I guess I could make that statement true. The point is, Krugman's last sentence here is not the statement of an empirical fact, so it does not count as evidence.

Then there is this:
And in a depressed world economy, any country running an artificial trade surplus is depriving other nations of much-needed sales and jobs. Again, anyone who asserts otherwise is claiming that China is somehow exempt from the economic logic that has always applied to everyone else.
Give me a break. Even if what Krugman claims earlier in the piece is true, China could only be accused of selling us stuff cheap, and lending us the funds to do so at a very low interest rate. Somehow this does not look like deprivation.

So what does Krugman want us to do about this "problem?" Well, apparently he wants a trade war. Great idea. Also,
One answer, as I’ve already suggested, is fear of what would happen if the Chinese stopped buying American bonds. But this fear is completely misplaced: in a world awash with excess savings, we don’t need China’s money — especially because the Federal Reserve could and should buy up any bonds the Chinese sell.
The world currently has a healthy appetite for US Treasury securities, which in part is why yields to maturity on all these assets are so low now. What happens if the Chinese, or anyone else for that matter, loses interest in our government's debt? Clearly, absent anything else happening, those yields will have to rise. But Krugman suggests that the Fed can buy the Treasuries that the rest of the world is unloading, thus monetizing the debt. This would of course result in more inflation, which would amount to an implicit default on our nominal debt, which I assume is what Krugman wants. This of course would make the world even more skittish about holding our Treasury securities.

Krugman either believes this stuff or he doesn't. In either case it is disturbing. If he doesn't believe it, then what's his agenda? Why the China-bashing? I expect this from politicians, but not from a Nobel-Prize winning trade theorist.


  1. As bad as the Krugman article is, this post is even worse. Probably even amateurish. To say that currency pegs don't affect exports defies both logic and historical evidence. The PBOC has trillions of dollars worth of reserves, and every time they allow the yuan to rise, the PBOC takes a huge loss on their balance sheet. Why do they do this? Just to please all their buddies on Capitol Hill? I think not. China does subsidize its exports via buying dollars (exporter's profits) for more than their trade-weighted worth, and the PBOC itself knows that it's a losing deal that they'll eventually have to take a loss on in the future (thus the revaluation). The FDI restrictions are important to the PBOC in order to keep the yuan weak (although there is recent news that they will begin to relax these, and this comes short after announcing the yuan revaluation that everyone was looking for. Coincidence? I think not). They subsidize their exports to bring their economy to full employment and use their financial system to control their citizens' savings to finance massive amounts of infrastructure, and all this is the strategy of state capitalism to transition into a more developed economy.
    Let me make a suggestion: you and Krugman should go on a date to China. You both can yell at each other, and work out your differences, and at the end of the day you two might learn a thing or two about basic Chinese economics.

  2. Sorry, I can't agree with your post either.

  3. A basic principle of international economics is that a country cannot "subsidize" its exports through what are essentially monetary policies determining its nominal exchange rate.

    No, that isn't a basic principle of international economics.

  4. Anyone say "trilemma"?

  5. "compared to the high domestic real interest rates in China."

    Perhaps I am misreading your post, but I don't see high real rates in China. Please can you clarify?

  6. Anonymous 1: I have actually spent some time in China, and learned a few things while I was there. China's government may indeed have some inefficient policies, but I think it is none of our business.

    Anonymous 2: Which part?

    Kevin: It's not a principle, it's a theorem, or a lemma? It's not basic? It's wrong? Explain.

    Last anonymous: What I mean is that, because of the inefficiencies in domestic financial intermedation in China, there are high real-return investment projects that go unfunded. In a world where, say, foreign banks had free reign in China, savings would flow into China, not out.

  7. On undervaluation, take a look at: Rodrik (2008), Brookings Papers; Berg and Miao (2010), IMF Working Paper; MacDonald and Vieira (2010), CESifo Working Paper; and, of course, the literature about FEERs and REERs.

    On undervaluation and growth see the previous three papers (BTW, Rodrik has both evidence and a model), plus Eichengreen (2008) "The Real Exchange Rate and Economic Growth" and Johnson, Ostry, and Subramanian (2007), IMF Working Paper (among the many others). They all think that the channel through which undervaluation affects growth is via trade.

    Hope it helps. Maybe Krugman's argument is wrong, but it was not completely out of the blue.

  8. It's not a principle, it's a theorem, or a lemma? It's not basic? It's wrong? Explain.

    Krugman did explain: “in effect, China is taxing imports while subsidizing exports.”

    You claim that China can’t be doing that, because it’s impossible. Actually it’s not only possible, it’s quite easy. (Admittedly it would be difficult for a less authoritarian regime, which would surely have to contend with a black market in foreign currency.) So your ‘principle’ is akin to the ‘principle’ that rockets don’t work in outer space because there’s nothing to push against.

    I suppose it’s possible you have in mind some model in which taxes on imports and subsidies on exports have no real effect, because prices of non-tradables and wages adjust, leaving incentives unaffected. But it’s up to you to say what your model is and argue for its validity. I know of nothing Krugman has written on this subject which conflicts with standard textbooks such as Krugman and Obstfeld’s International Economics.

  9. Kevin:

    So the prices of things that China buys and sells on world markets are fixed permanently in units of Chinese currency? Is that your argument?

  10. trader:

    Read for example Section 1: "Is the Real Exchange Rate a Policy Variable?" in Eichengreen's "The Real Exchange Rate and Economic Growth." That's pretty much consistent with what I am saying here.

  11. "Which part?"

    Is that a serious question? I'm not going to explain this stuff to you but I think you should think about EU countries in the ECU and some issues that happened in South America.

  12. That piece is consistent with Krugman's statement.

    "China is deliberately keeping its currency artificially weak.

    The consequences of this policy are also stark and simple: in effect, China is taxing imports while subsidizing exports, feeding a huge trade surplus."

  13. Anonymous 2: As serious as I can be. What about the EU and South America?

    Last anonymous: "That piece" being which?

  14. Scott Sumner has also been defending China from Krugman:

  15. So the prices of things that China buys and sells on world markets are fixed permanently in units of Chinese currency?

    No, the subsidy/(tax) which Chinese exporters receive (and importers pay) is determined by the Chinese government's exchange rate policy. If France slaps a tariff on Japanese TV sets that doesn't fix the price in Euros 'permanently' or even temporarily. But it surely hurts Japanese exporters and French consumers.

  16. Rajan has a nice post a while ago on global imbalance:

  17. In the short run it depends on the Marshall Lerner condition. Over the long term, prices have to adjust.

  18. Last anonymous:

    Exactly. Marshall-Lerner assumes changing the nominal exchange rate changes the real exchange rate in the short run, and then the effect on the trade balances depends on elasticities. I'm saying that we're talking about tradeables. The prices of Chinese exports are determined on world markets in US dollars, and the prices adjust immediately. For Chinese imports, maybe there is some market segmentation, so importers don't change their prices immediately, but maybe not. In any case, in the long run the prices have to adjust. Krugman seems to be telling us that the prices never adjust.

  19. "Krugman seems to be telling us that the prices never adjust."

    PK's blog is political, not economic. What he's telling us - at least on my interpretation - is that the time required for the adjustment (via internal prices rather than the nominal exchange rate) is 'too long' for the Democratic political establishment he represents. Remember he's an unreconstructed Keynesian, and you know what Mr. Keynes said about the long run.

    Regards (and thank you for your earlier clarification), 'the last anonymous'.

  20. China receives US dollars from its huge nets exports to the US and then mandate its citizens/corporations to exchange these US dollars for Yuan at a given exchange rate if they want to do business in China -these Yuan being created out of thin air by the Central Bank. The People's Bank of China then turn around and buy US Treasuries with the US dollars, and the Yuan is circulating in the domestic economy as the national currency. Because of this currency swap at a fixed exchange rate, the total assets on the balance sheet of the People's Bank of China amount to a gargantuan 3.4 trillion US dollars.

    You aim to destroy Krugman with this post and all you do is covering yourself in ridicule with your story that financial intermediation in China is inefficient which would supposedly explains why China buys US treasuries. How long have you studied China's banking system to come up with such statement or are you just making stuff up on the go?

  21. "In any case, in the long run the prices have to adjust. Krugman seems to be telling us that the prices never adjust."

    This is the point. If prices adjust sluggishly, a change in the nominal exchange rate will affect the terms of trade and hence the trade balance. How long does it take to adjust? I don't know, trade economists have different opinions about this (Obstfeld & Rogoff, grad textbook has a nice treatment of the issue). But it is definitely conceivable that affecting a nominal variable has real effects for some time and so Krugman's argument is not so far-fetched as you are making it.

    Which is different from saying that I agree with the whole thrust of his op-ed, or, more in general, with his "the more goverment, the better" general philosophy.

  22. Bils and Klenow report that the average price-setter resets every 4 months. That doesn't seem like such a long time. Further, that's just an observation about average behavior. It doesn't tell you what these price-setters do when there is a policy shock.

  23. I'd never thought I'd be defending Krugman, but I guess there always has to be a first...

    "In any case, in the long run the prices have to adjust."

    But when is the long run? And what happens from now until then?

    "China could only be accused of selling us stuff cheap, and lending us the funds to do so at a very low interest rate. Somehow this does not look like deprivation."

    But maybe looks are deceiving? That is, maybe your description of the situation is not complete and lacks secondary effects which are real and harmful (people thrown out of work who had jobs selling stuff that are now sold more cheaply).

  24. Stephen thanks for the reply. I have actually read section 1 in Eichengreen, as well as the other sections and the references that I mentioned. Barry says “My reading of the evidence is that the real exchange rate matters. Keeping it at competitive levels and avoiding excessive volatility are important for growth”; and, again: “the real exchange rate is a relative price and, as such, it is not under direct control of the authorities. But it can be influenced by policy.” Personally, I think that a country that maintains a fixed peg, has in place a system of capital controls and limited currency convertibility, and is endowed with “unlimited supply of labor” (in the sense of Lewis, 1954) that can prevent wages in the tradeable sector from rising, can actually maintain a competitive (undervalued) currency “beyond the very short run”.

    Anyway, to tell you the truth, the point of my comment was quite different. You seemed to be unaware of the fact that many people --- Dani, Barry, Simon Johnson, people at the IMF, people at the World Bank (I forgot to mention their series of papers), and many others --- have been working on these issues. There is a lot of stuff on undervaluation as well as on undervaluation and growth. Krugman did not “invent” an argument.

    As I often read his blog, I understand that he can be quite irritating, especially with his frequent accusations of ignorance about basic economics. But you maintain a wonderful blog and my suggestion --- as a passionate reader --- is to try to avoid doing his same mistake.

  25. Thanks for your answer, Steven. Two things here. First, resetting a price is not the same thing as completely absorbing the nominal shock. Adjustment could still be gradual.

    Second, my point is similar to Trader's point above. There are quite a few international economists who believe that the nominal exchange rate matters. You can argue that they implicitly rely on price-stickiness which does not seem to show in the data. And so that their beliefs is not warranted. But this is an argument to be made, not a basic principle of international economics.

    I also agree with the last Trader's paragraph.

  26. You lost me when you wrote that your bother "complains when the value of the Canadian dollar goes down in terms of US dollars". His income in US dollars exchanges for more Canadian dollars, thus increasing his income to the extent his expenses are denominated in Canadian dollars. Why does this make him sad?

  27. Yes, he would be happy wouldn't he. I was in a hurry and made a sign error. Let's try this again. The renminbi appreciated from 2005 to 2008, and after that it has been essentially pegged against the US dollar until now. Krugman's claim is that the renminbi is "undervalued" and "artificially low" relative to the US dollar. He knows what the exchange rate "should" be, and it's higher than where it is now. Thus, it's not a case of: country devalues, thus moving relative prices in the short run. Of course, Krugman thinks that if the renminbi appreciates that the real exchange rate moves in the same direction. Now the bizarre view here, which I think is a very poor basis for making economic policy, is that the countries of the world are competing for jobs through their nominal exchange rate policies. The view is that the Chinese are stealing our jobs because they will not allow their exchange rate to appreciate. According to this view we therefore put pressure on them to do so, but if they retaliate by unloading our debt, then we have the Fed buy the debt, thus depreciating our exchange rate and stealing the jobs back. Maybe this makes sense to you, but I think it is nuts.

  28. It does indeed make sense to me. If China keeps accumulating dollars, thus keeping its currency lower than if it doesn't do so (and yes, this is obviously what is meant by keeping it artificially low, scare quotes or not) then they can maintain a large trade surplus for a long time. We get lots of cheap Chinese stuff in exchange for many of us being unemployed. You can argue about whether this is a good tradeoff, but this is a policy decision, not one that is as clear cut as you imply with all the histrionics about Krugman's position being "nuts" and "bizarre".

  29. I gave you a story about why China is running a current account surplus, which is about savings, investment, and inefficient domestic financial intermediation in China. Krugman's idea appears to be that the current account surplus tells you the exchange rate should appreciate. From my point of view, if Krugman thinks that the current account surplus should be zero in China (not clear why this should be so either, in general), then he should be convincing the Chinese to do something about their inefficient intermediation problem - e.g. opening up competition with foreign banks. Now, the large quantity of Chinese savings that leaves China via the current account surplus makes world real interest rates lower than they otherwise would be. Explain how that makes us unemployed.

  30. I agree that Krugman is making irresponsible oversimplifications. The nominal exchange rate, by itself, is not the issue, because if it has an influence, it is only in the very short-run. I think the issue is whether a broader policy to facilitate exports at the expense of imports that commonly includes a nominal exchange rate peg, along with financial repression, etc. is "unfair". [I wouldn't have put it like Williamson- I do think that conscious policy choices and not just incompetent bankers are responsible] To my mind, the answer is that the question is poorly posed- on one level, if the Chinese want to keep bankrolling a US consumer spending spree (directly through selling goods cheap and indirectly by lending money cheap), hey, if anyone should think its unfair, it is really the ordinary Chinese.

    The problem is that US institutions (labour markets in particular), are not set up to deal w/ the sudden inflow (and eventual outflow, one hopes, although I have always dreamed of a permanent retinue of Asian servants) of relatively (to the long run) underpriced consumption.

    What Krugman and his ilk are really upset about is not China's nominal exchange rate, but the inability of the US' institutional set-up that he quite likes to deal with changes posed by China's rise. Being a US Liberal of the 1960s/1970s variety, and thus wedded to the way things are, it is far easier to blame the renminbi.

  31. "I wouldn't have put it like Williamson- I do think that conscious policy choices and not just incompetent bankers are responsible"

    No, that wasn't what I meant. The banking system is dominated by a few large state-owned banks. The banks are essentially part of the government, if you like, so the inefficiency is all about government and policy.

  32. A very stimulating post, and lots of good exchanges that follow --- including Dr. Williamson's willingness to reply to his critics.

    1) That said, it's Catesby who is the first to point to the short- or mid-term problem of China's exchange rate policies --- which policies benefit American consumers at the expense of Chinese consumers (rich and poor alike), and make American interest rates lower than otherwise . . . this time, at the expense of Chinese savers who get ludicrously low returns on their financial investments.

    2) That major problem that Catesby highlights?

    In a huge Continental-size country, with lots of diverse goods and service industries and a differential need in a post-recession recovery for jobs, China's extraordinarily low level of domestic consumption even by Northeast Asian standards (Japan, South Korea, Hong Kong, Taiwan, and further south Singapore) --- something like an unheard of 36% of GDP recently --- has accelerated through its large bilateral trade-surplus with the USA the pace of labor-market adjustments: in particular, by the surpluses’ impact on where American jobs have been destroyed in industries that became uncompetitive with Chinese (and other imports). Without that impact, our national labor markets would have had more time to adjust to these shifts.

    3) What follows? Well, if labor markets adjusted without frictions, with little time-lags to large shocks in shifting comparative advantage, the overall national rate of unemployment since December 2007 would have been lower, right? And possibly much lower, no?

    Enter the reality of our labor markets --- which have traditionally been unusually flexible. In particular, as comparative advantage of our tradable industries has shifted quickly, the result is that the geographical location and labor-skills needed for our growth industries and their firms seem mismatched with a large chunk of the unemployed since December 2007.

    4) None of this denies that the overall rate of unemployment in the USA will adjust sooner or later and will be overwhelmingly determined by American market influences and government policies. But the distribution of job-creation across our goods and service industries has surely been affected by the huge surge of Chinese imports and our trade deficit with China itself since 2000.

    In that year, the Chinese trade surplus with us was about $90 billion . . . the same as Japan's. In 2007, our bilateral surplus with Japan was about the same, but China's surplus had surged four-fold. Our consumers surely benefited --- mostly the low-wage income earners among them; and so did those households looking for credit to buy consumer durables thanks to lower interest rates as a result of China's Central Bank buying our Treasuries, just as business firms could invest at lower rates.

    But the resulting job-market dislocations seem now to have aggravated what you yourself have talked about: a mismatch between the high number of the unemployed and jobs requiring skills they do not easily have . . . or, if they do have them (a certain percentage anyway), the collapse of the prices of any housing they bought earlier in the boom period of the last decades makes it harder for them to move quickly to the areas of the country where jobs are being created steadily.

    5) In short, when labor markets don't adjust quickly for whatever reasons, the normal processes of "creative destruction" (to use Schumpeter's term for how capitalist markets adjust to clustered changes in innovative technologies of a basic sort) have been partly clogged by China's huge mushrooming surge in exports between 2000 and now 2010 --- with a brief slowdown and drop in 2008 and early 2009 --- that have aggravated, it seems, our overall post-recession unemployment rate, whether or not you regard it as a short- or mid-term problem that will right itself in the "long-run."

    Michael Gordon, AKA the buggy professor