How Seriously Should the U.S. Take China's Economic 'Nuclear Option'?

China's 'nuclear option' threat of economic retaliation against Trump's trade war is more akin to a toy gun.

Photo Credit: William Potter / Shutterstock

Once the trade war rhetoric heated up, it was only a matter of time before someone raised the possibility that China would begin selling its large holdings of U.S. Treasuries, causing “a real rout” in the bond market, according to the former chairman of Morgan Stanley Asia, Stephen Roach. Reuters followed up with a provocative headline: “China, holding Treasuries, keeps ‘nuclear option’ in U.S. trade war,” suggesting that China was holding its “largest import”—namely, its huge cache of $1.17 trillion of U.S. Treasuries, in reserve, should the trade war substantially escalate from here.

It sounds very serious, but are these claims about China actually true? Well, not according to the Chinese… yet. Vice Finance Minister suggested that China was “a responsible international investor,” and had no intention of exercising the nuclear option, not even to give the U.S. “a bloody nose,” as Washington itself has contemplated with a Chinese ally in some of the racier moments of Trumpian diplomacy. On the other hand, as CNBC reported, “the Chinese ambassador to the U.S. Cui Tiankai responded to a question on Treasury bond purchases simply by saying that, ‘If the other side makes a wrong choice, we have no alternative but to fight back.’”

Of course the problem with a “nuclear option” is that once it is exercised, all leverage is gone… along with everything else. They are like lawyers, as Danny DeVito memorably said in “Other People’s Money”: “[Y]ou use them and they fuck everything up.”

Colorful threats aside, does China actually have the economic equivalent of a nuclear deterrent via its U.S. bond holdings? Certainly, Beijing has every interest in making the threat sound as scary as possible as it gives them leverage. Call it the Chinese version of The Art of the Deal. And it sounds credible when you have figures like Stephen Roach suggesting that the bond market would be overwhelmed and prices would crash were Beijing to initiate large-scale liquidation of its U.S. bond holdings (which, if true, might be a classic case of China cutting its nose to spite its face).

But does the threat stand empirical scrutiny? As the head of currency trading at Brown Brothers Harriman, Marc Chandler, has noted, “From June 2016 through November 2016, China’s Treasury holdings, according to U.S. data fell by $200 bln, which was 15% of their holdings. What happened to the U.S. 10-year yield (as a rough and ready metric of the impact), you ask? It was virtually unchanged.” Not only that, but the Chinese currency, the renminbi (RMB), appreciated against the U.S. dollar, hardly an optimal outcome if one is trying to mitigate the impact of an adverse trade shock.

Recall how China gets its dollars (and Treasury cache) in the first place. They are not an “import,” obviously, but a byproduct of a market transaction. China sells, say, $1bn worth of electrical machinery to the U.S. and has those dollars credited at the Bank of China’s account at the Federal Reserve. And, as the economist Randy Wray argues: “The Bank of China rationally prefers to earn interest on dollar holdings, so these are converted to U.S. treasuries. This is nothing more than a balance sheet operation on the books of the Fed: Bank of China reserves at the Fed are debited and Bank of China treasuries are credited. There’s no net flow of dollars to the U.S. Treasury.”

This whole notion that China sneezes at the bond market and the U.S. catches pneumonia is predicated on a rather archaic model of capital flows in which bonds and dollars are said to “flow” across borders like ocean waves, creating a tsunami-like impact as the flows intensify. That image may have been more reflective of reality during the days of the 19th-century gold standard, but in today’s computerized world of modern finance, dollar balances (and/or Treasuries) are electronic entries on the balance sheet of the Federal Reserve. So if China wants to sell its U.S. Treasuries (which are held in a Federal Reserve “securities account,” a fancy term for “savings account”), a bank officer at the Fed presses a button, the Treasury holdings are debited from the securities account and the corresponding dollar holdings are immediately credited to a “reserve account,” which is the Federal Reserve equivalent of a “checking account.” In fact, as the economist/portfolio manager Warren Mosler describes, this is routinely how all U.S. debt management is done: “The Fed removes dollars from savings accounts and adds dollars to checking accounts on its books. When people buy Treasury securities, the Fed removes dollars from their checking accounts and adds them to their savings accounts.” In other words, capital doesn’t “flow.” And from the perspective of the U.S. government, if China’s dollars simply sit in the reserve accounts, it just means issuing fewer bonds, which means paying out less interest to Beijing.

Once China gets its dollars from an export sale, it has several options: it can let the resultant dollars accumulate in their reserve/checking account at the Fed. Or it can use the dollars to buy other U.S.-dollar denominated assets (e.g., real estate, a U.S. company, commodities, etc.), which would mean spending the money in the U.S. economy, which would also be good for the American economy. Alternatively, it can sell the dollars and swap them into another currency, such as the euro or back to the RMB itself (which was done in the latter half of 2016 when China was trying to support its currency against capital flight).

Stephen Roach’s characterization of what happens in a bond market transaction, therefore, is mistaken. In a non-gold standard world of free-floating fiat currencies, what becomes the equilibrating factor here is not the interest rate (which used to be impacted by the physical flow of gold going inside and outside a country), but the external value of the dollar vis a vis other currencies. If Beijing neither retains its dollars, nor U.S. Treasury holdings, and instead opts to sell them, then, all other things being equal, it is possible that other private portfolio preference shifts could well be temporarily influenced by China’s actions (much as the decisions of a wealthy, well-known art dealer at an auction might well impact the actions of other participants). “All other things being equal” is the key phrase here, even though all other factors seldom are equal. These shifts in portfolio preferences could in turn put downward pressure on the dollar’s external value, or the price level of U.S. bonds. But not necessarily (it didn’t, for example, the last time China engaged in large-scale liquidations of its bond holdings). Of course, in the event that the dollar did decline, it would have the happy byproduct of mitigating America’s trade deficit, as it would boost the competitiveness of the country’s exporters, which was the starting point for the current outbreak of protectionist pressures in the first place.

No less than Nobel Laureate Paul Krugman puts paid to the notion that China truly possesses a dangerous nuclear option:

“[W]hatever China’s motives, the Chinese wouldn’t hurt us if they dumped our bonds—in fact, it would probably be good for America.

“But, you say, wouldn’t China selling our bonds send interest rates up and depress the U.S. economy? I’ve been writing about this issue a lot in various guises, and have yet to see any coherent explanation of how it’s supposed to work.

“Think about it: China selling our bonds wouldn’t drive up short-term interest rates, which are set by the Fed. It’s not clear why it would drive up long-term rates, either, since these mainly reflect expected short-term rates. And even if Chinese sales somehow put a squeeze on longer maturities, the Fed could just engage in more quantitative easing and buy up those bonds.

“It’s true that such actions could possibly depress the value of the dollar. But that would be good for America!”

Even if the Federal Reserve has no inclination to engage in a further round of quantitative easing, it has other available tools in its policy box, especially since it has started to pay interest on excess reserves (IOER), as Rob Parenteau and I outlined in a 2010 article:

“[T]he Fed can tell everybody that they are renormalizing the fed funds rate and take the IOER up to 100bps. Note, the Fed does not need to remove any reserves to do this—they can just do it administratively. That’s how the IOER works—it severs the link between reserves in the system and the target policy rate, right?

“Then, if the bond gods don’t rally Treasuries on the Fed’s efforts to renormalize the policy rate, Mr. Bernanke calls up Bill Dudley (President at the NY Fed) and gives him instruction to buy all the 10 year UST on offer until the 10 year UST yield is down to, oh, say 3.5%. It is an open market operation, which the Fed performs all the time. They won’t have to call it QE, but it is in effect the same thing.”

If China’s sales were to cause long-term rates to go above a certain level, the Federal Reserve could instruct its open market desk to buy these Treasuries to counter the sales’ impact, if necessary. To quote Paul McCulley, formerly of PIMCO (the world’s largest bond fund manager, which means the group knows a thing or two about how the bond market operates): “any market induced—foreign or domestic-driven—upward pressure on U.S. intermediate or long-term interest rates would/will be limited by the leash of the Fed’s... anchoring of the Fed funds rate. ... Put differently, there is a limit to how steep the yield curve can get, if the Fed just says no—again and again!—to the tightening path implicit in a steep yield curve.” Yes, that’s not how a “free market” is supposed to work, but by definition, a market isn’t totally “free” if ultimately dominated by a monopolist supplier dollars and treasuries. That the Federal Reserve may choose not to exercise that monopoly power is a political decision, not a capitulation to market forces in which the central bank’s actions are overridden by an onslaught of Chinese bond sales.

To that extent, the famous, all-powerful “bond market vigilante” is in reality an astute arbitrageur, who ultimately can exploit any differential between the Fed’s stated objectives and market pricing, so long as the Fed backs up its intentions credibly, as McCulley suggests. Yes, it is true that there have been examples where markets have overwhelmed governments, but these have occurred in instances where the government (or its central bank) was trying to control an arbitrary price level in which it was not the monopoly issuer (such as would occur, say, with a currency pegged to the price of a foreign-denominated currency and therefore viable for only as long the government has sufficient stocks of said foreign currency). Or to put it another way, governments/central banks can control price or supply, but cannot control both simultaneously. The Fed has an unlimited war chest of dollars if it so chooses to defend a particular price point on the long-term bond. So does the Bank of Japan in regards to yen-denominated assets (failing to understand this basic fact, market “experts” have been wrongly predicting disaster for the Japanese bond market for eons). Or the Bank of England in regard to sterling. This is what is meant when a government is described as being truly “sovereign” in regard to its own currency. It means that it is the sole issuer of its own unit of account and nobody else.

In reality, then, China doesn’t possess a “nuclear option” here, because “all the dollars the Chinese have come from the U.S. There is no net supply of dollars from China.” If you don’t own the nukes, the threat, as the economist Dean Baker has poetically phrased it, becomes “an empty water pistol pointed at our head.” The Chinese have played this card effectively in the past because the U.S. has never called their bluff. Beijing is counting on the current market turbulence to pile on the pressure. Additionally, as Japan did in earlier trade wars with the U.S., China has learned to mobilize U.S. agricultural interests, and low value added industries against the defenses of our manufacturers. Hence, the reason for the retaliatory tariffs on imports of U.S. pork, leather and scrap metal. This hasn’t proved as effective as in the past so far, as Trump has responded by upping the ante by instructing the USTR to put together a plan for $100bn in additional tariffs against Beijing. In the meantime, China’s policymakers continue with an explicit import substitution policy, targeting industries currently dominated by American companies, even as the latter face constant threats of intellectual property theft and domestic joint ventures in which they are compelled to transfer valuable technology to Chinese firms, to get access to a domestic market in which China remains the favored player (actively aided and abetted by the government, as the latest USTR Report to Congress documents). It’s a highly effective policy, as it enables China to short-circuit the move up the technology curve and avoid the high costs normally associated with research & development. This is not “free trade” as the textbooks describe it, but a highly effective mercantilist strategy, one which Washington has allowed to be sustained on the basis of a bogus threat from a paper tiger.


Marshall Auerback is a market analyst and commentator.