Morgan Stanley Misreads China
Last week Morgan Stanley’s China economist, Qing Wang, circulated a Forbes article keyed off his quotes regarding China’s stimulus spending. The article and his quotes contain some deeply flawed reasoning, which I think is worth highlighting because it illustrates some commonly held misconceptions, ones that seem especially prevalent within China. I invite you to read the whole article, but you can catch the gist of Wang’s argument here:
Since China has to invest its record foreign reserve holdings of $2.1 trillion anyway, building more roads and bridges at home will likely generate better returns in the long run than buying U.S. government debt, Morgan Stanley says . . .
. . . Thus the question becomes simpler and clearer: whether China should use its rising reserves to build more bridges, roads, and railways; or to buy more US government bonds – which is what China’s central bank does with most of its vast foreign exchange holdings.
“In view of the negative consensus outlook for US government bonds, we think that infrastructure investment is a better alternative for deploying China’s savings, especially when the cyclical conditions warrant a strong boost in domestic demand in a relatively short period of time,” Wang concluded.
For China, building bridges is a better bet than buying bonds.
There are two main problems with this line of reasoning. First, it presumes that China is reallocating its savings from U.S. Treasury bonds to domestic infrastructure investment. But China’s foreign currency reserves are not fungible in this way. When the Chinese choose to accept dollars for their exports, they receive dollars, and those dollars don’t just disappear. The Chinese can sell them to someone else who wants dollars (which would decrease the value of the dollar relative to the RMB or whatever other currency the Chinese preferred to possess). Or they can buy goods, services, or assets denominated in U.S. dollars (usually imports from or property located in the U.S., but also global commodities such as oil). Since China is generally unable or unwilling to allow these two actions, and their consequences, the Chinese government must buy the dollars and hold them itself, mainly in the form of U.S. sovereign securities.
But the point is, the portion of Chinese savings that arises from consuming less than it produces, and selling the excess abroad (in order maximize growth), is denominated in dollars (or other foreign currencies) and must be spent in dollars. If China doesn’t want those dollars, or doubts their value, it has a simple option: demand more dollars in exchange for same goods, i.e., appreciate the Renminbi. But China doesn’t want to do that, because it want to keep selling, so it keeps piling up more and more dollars.
Wang’s point would make sense if China’s stimulus was being spent to buy goods and services from the U.S.–say, bulldozers to build roads or MRIs for new hospitals. In that event, China could take saved dollars invested in U.S. Treasuries and reallocate them to buy U.S. imports. But that is not what we see happening. And China’s recent adoption of “Buy China” regulations for its stimulus package essentially guarantees that we won’t.
China’s stimulus investments are savings, to be sure. They have to be funded, either by issuing government bonds, or raising taxes, or through inflation (which is effectively a silent form of taxation). But this is new savings, on top of the dollar-denominated savings accumulated from its trade surplus. China isn’t reallocating savings more effectively, as Wang would have it. It is adding to savings, at a time when most economists believe it should be decreasing its savings rate.
The second problem with Wang’s analysis is it assumes that, just because U.S. Treasuries may not be the optimal investment, anything must be better. Right now, the nominal yield on a 10-year Treasury note is about 3.6%. But the criticism of China’s grander stimulus projects–one I can validate from my recent travels around the country–is that many of them may end up being “white elephants” that produce negative returns on investment. I recently visited a brand new state-owned steel mill in Liaoning Province that, after a year of operation, had yet to fire up its furnace because of global overcapacity. Surrounding that plant, as far as you could see, were leveled construction sites where a huge new city was being built to accommodate roughly 3 million people. And this was just one of several massive port and industrial projects in the immediate vicinity–China’s stimulus money at work. Visually impressive, yes. It provides jobs, too. But I’d never invest. I’d far prefer to put my money in safe, low-yielding Treasuries.
Wang raises two counterpoints that have some validity. First, he says these projects may have social benefits–positive externalities–that will enhance China’s overall productivity, and therefore cannot be evaluated purely by looking at commercial returns. True, but I’m not wholly convinced, particularly if the negative commercial returns involve substantial loss of capital. Purely social goods like water purification, parks, and health clinics may produce more benefits than they consume as cost centers. But 50% of China’s RMB 4 trillion approved stimulus is earmarked for transport and other infrastructure projects–railroads, highways, airports, dams, power plants–that really should pay for themselves. If they don’t, you have to wonder if they are viable. China does need infrastructure, particularly in its less developed interior, but I’ve also seen plenty of examples of political vanity projects and other boondoggles that have never justified their costs. Shanghai’s maglev train, which runs from the airport to the middle of nowhere, and thus carries virtually no passengers (although it runs constantly), is merely the most glaring example.
Second, Wang hints at Chinese fears that inflation in the U.S.–brought on, ironically, by rampant stimulus spending, which Wang now seems to assume is unproductive–might undermine the value of the dollar. If that were the case, real after-inflation returns on U.S. Treasuries could be negative for China. It is true that such fears partly account for the market demanding higher yields on 10-year Treasuries, up from 2.1% in December 2008. But yields are still lower than they were a year ago, before the real crisis broke. That implies that most global investors still find Treasuries a relatively attractive option, particularly as a “safe harbor” for capital preservation.
I will admit that the second point, about the relative risk-adjusted returns of U.S. Treasuries versus China’s infrastructure investments, is arguable. It really amounts to a judgment call on U.S. finances and Chinese state planning. But the first point, regarding the non-fungibility of China’s dollar-denominated savings, is far more important. There is no “better way” to use China’s foreign currency reserves, as long as it remains wedded to an export-dominated growth model. They cannot be “reallocated” to fuel purely domestic growth. The only way to unlock China’s savings is for China to open its economy to buy and invest more abroad. “Going it alone” is simply not an option.