Reply to Andy, Nick, et al, on Krugman
Because I teach several courses today, I don’t have time to respond to all the comments right now. But I skimmed those after my Krugman post and I really think almost everyone is looking at these issues in the wrong way. People are confusing historical claims with theoretical presumptions, and making distinctions between “natural” and “artificial” that are themselves . . . well, totally artificial and arbitrary.
Start with this statement by Paul Krugman:
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.
That could mean several things:
1. Every time there is a trade surplus, there is an undervalued currency.
2. Every time there is an undervalued currency, there is a trade surplus.
3. The more undervalued the currency, ceteris paribus, the bigger the trade surplus
It seems to me that 99% of Krugman’s readers would assume he meant #1 or #2. But obviously both are wrong, even my critics seem to agree on that. (As I’m sure Krugman would.) After all, they define “undervalued” on the basis of government intervention in the foreign exchange market to depress exchange rates below their natural value. So to defend Krugman you have to assume he meant #3. And I imagine he did. But what a misleading way to make a point. He would then merely be saying that there is a theoretical presumption that undervalued currencies make surpluses bigger. But then why suggest there are no historical counterexamples? There is no real world evidence I could cite that would disprove that assertion. No matter how many countries I found with CA deficits, that were also depressing their currencies by buying foreign exchange, he could say that the action was still, ceteris paribus, making the CA deficit smaller. When you challenge people to find one real world example that would refute your argument, as Krugman does in that sentence, there is an assumption that he is asserting something more than a tautology. And I’m sure his readers read it that way. “Look, China has a big surplus, find me one of those that wasn’t caused by an undervalued currency.” If his defenders’ interpretation is correct, then his call to search the world for a counterexample is essentially meaningless.
An even bigger problem occurs when people try to differentiate between “good” surpluses due to “natural forces” and bad surpluses that are “artificial.” This is associated with confusion about the role played by China’s exchange rate peg. Here are some key points:
1. If China stopped pegging the yuan and let it float, but continued buying $100s of billions in foreign exchange every year, then China would continue to run large CA surpluses. The root cause is high Chinese saving (relative to investment) and the currency is merely the transmission mechanism. As an analogy, the root cause of less gasoline consumption in 1974 was OPEC producing less, and the transmission mechanism was the high price. OPEC didn’t even need to peg the price, just produce less.
2. If China stopped pegging the yuan, and stopped buying dollars, they would still run a huge CA surplus as long as the Chinese government found some other way to save a lot. Suppose they set up a Norwegian-style sovereign wealth fund, and bought $100s of billions worth of German, French and British equities each year. Nothing from the US. Chinese aggregate saving would still exceed aggregate investment, implying a big CA surplus. And the Chinese would not be interfering at all in the forex markets, as the term is usually defined.
3. Ah, but some of you laissez-faire types (who strangely support Krugman) will say that’s not pure enough, not virginal enough. The Chinese government is still involved in all that saving. It’s artificial. You’d say “saving should be done by the free markets, not governments.” You’d say only in that case is a CA surplus OK. Otherwise the exchange rate is still at an artificial level. Evil countries like China and Norway must be punished for all their governmental saving.
4. OK, the Chinese government stops intervening in the foreign exchange markets, and stops saving of any kind. But instead they set up a Singapore fiscal regime, and force all workers to set aside 31% of income into various private forced savings accounts. China would still save more than it invests, and still run a huge CA surplus. Indeed Singapore’s CA surplus is much bigger in relative terms than China’s. Is this OK? After all the Chinese government is not saving or intervening in the forex markets. It is private citizens doing the saving. I can just see people responding; “No, that high saving is still tainted, still artificial. To be truly virginal, truly free of sin, the country has to save without any government encouragement.”
5. I give up. Sorry for being so ridiculous, but I’m trying to make the point that we are thinking about the entire issue in the wrong way. Seriously, at what point on the preceding list does the distinction between natural and artificial become at all meaningful?
What I find so bizarre about these “natural” and “artificial” distinctions is that in trade theory they are almost universally viewed as bogus. If Korea exports cars to US at a low price, when is it evil? When they directly subsidize production? When the government builds schools that train automotive engineers? When they provide cheap land for car factories? Economists usually roll their eyes at those distinctions. All that matters is whether the US gains from being able to buy low-priced cars from Korea. Similarly, if foreign CA surpluses really did hurt us we should oppose them whatever their cause. To be fair, Krugman has criticized Germany’s free market surplus, arguing the German government needs to save less. So he clearly understands the point I am making. Where he and I disagree is that I don’t think foreign CA surpluses hurt us by reducing NGDP, because I believe NGDP is determined by monetary policy. So I’m not upset with either China or Germany, and I certainly don’t draw any moral distinctions between the two cases.
Nick Rowe asked:
Think back to your earlier analogy, that if all countries intervene in forex markets, and buy each others’ bonds, it is like every country doing QE. Which is good. But if the Bank of China can buy US bonds, but the Fed can’t buy Chinese bonds (because China won’t allow it), then it’s asymmetric.
I’m not quite sure what Nick is getting at here. If he is saying that if each the 20 biggest countries bought $100 billion in bonds from the next country to its west, or to its east, it would be like a global QE, then I agree. Of course the exact same effect would occur if they each country bought $100 billion of their own bonds. If China doesn’t want to sell us bonds, we can weaken the dollar against goods and services (which is all we care about) by buying German bonds, or even US bonds. It makes no difference. Of course that’s assuming the transactions affect the money supply. But there is no necessary effect of Chinese forex purchases on the Chinese money supply, as I presume they sterilize them to prevent inflation. Only if China’s peg forced it to buy more forex than they felt comfortable holding, would the Chinese purchases be monetized.
Another reason the exchange rate is very misleading is that people focus on the nominal rate, which the Chinese government pegs. But it is the real rate that matters. It is true that China can influence the real rate, but only by adjusting the amount of government saving. Which against shows it is the saving/investment relationship, not the exchange rate, that is the key to understanding CA surpluses.
Andy Harless said:
The argument would be that currencies are undervalued whenever governments collectively engage in reserve transactions that result in a (sufficiently large) net increase in the availability of that currency in foreign exchange markets. (Obviously if the US were to offset China’s dollar purchases by buying yuan, this would not be the case.)
That seems like a pretty good argument to me: if the value of the currency is what the private sector would determine it to be in the absence of official reserve transactions, then it is fairly valued (on the assumption that private markets are efficient enough to give it a fair valuation). If net official reserve transactions are reducing the actual market value of the currency relative to that “fair” market value, then it is undervalued.
First of all they aren’t buying dollars, they are buying bonds. Yes, they are bonds denominated in dollars, but it would make no difference whether they bought bonds denominated in euros, or Swedish kroner, or stocks denominated in euros. All that matters is the impact on Chinese national saving. Again, there are lots of countries that have governments buying forex, which also run CA deficits. Whether you run a surplus depends on the saving/investment balance. I am not denying that governments can influence that relationship, and I am not denying the Chinese government has influenced it (as has our government, by massively discouraging saving), what I am denying is that the “official reserve transactions” affect the CA surplus more than some other form of government saving.
In the previous post I said:
“The standard argument is that trade surpluses (actually CA surpluses) are caused by excesses of domestic saving over domestic investment.“
Andy Harless responded:
That doesn’t sound right to me. The level of the CA surplus is simultaneously determined with the levels of saving and investment, but the levels of one do not cause the levels of the other. If the quantity that a country wishes to invest and the quantity it wishes to save are both fixed (inelastic) at certain levels, then those quantities will determine the CA surplus, but they will do so largely by affecting the exchange rate. (For example, the excess saving will bid down domestic interest rates, thus making the domestic currency less attractive and causing it to depreciate, so that domestic goods become more attractive relative to foreign goods, thus inducing a CA surplus.) The exchange rate is still the proximate cause of the CA surplus. There is no “immaculate transfer.”
This is like the oil example. The OPEC cutback in production was the root cause of Americans’ driving less. If you want to call higher oil prices the “proximate cause,” that’s fine. I’d call it the transmission mechanism. Suppose OPEC had gradually increased prices in the 1970s and 1980s by making “National Parks” out of land they had previously intended to drill new wells in. That sounds very environmentally conscious, doesn’t it? But it would have had the same effect on prices as an “artificial” order to the oil companies to produce less. I’m saying those distinctions are meaningless. It doesn’t matter why oil production fell, only the fact that it did. Norway gets praised for its high national saving rate; “They are so wise to put their oil money into a sovereign wealth fund, unlike all those less civilized countries.” You know the things people say. But when China does the same thing, they are viewed as being “sinister.” Yes, the CA surpluses are probably not in China’s interest (unlike Norway), but that’s hardly a reason to punish the Chinese people.
So what am I missing here, as everyone seems to disagree with me?