Currency Reserves

This gathers the six post series I did with Concentrated Ambiguity into a single easy accessible post.

Taiwan's central bank—and Taiwanese households—have taken on an awful lot of foreign currency risk over the years …

Based on the profits and losses disclosed by Taiwan’s central bank, it appears that its true FX exposures exceed its disclosed foreign exchange reserves by USD 130bn, and perhaps by as much as USD 200bn.

Who is on the other side of the massive ($250 billion) hedging need of Taiwan's life insurance industry? The local banking system covers at most a quarter of the life insurers hedging need, and foreigner investors provide less than that... 

Taiwan’s central bank, unlike most central banks, doesn’t disclose its position in FX derivatives. It really should. There is good reason to think its undisclosed exposure is quite substantial.    

I have a new Policy Innovation Memo that recommends three changes to U.S. currency policy, and specifically, three changes to the U.S. Treasury’s Foreign Exchange report: 1. The Foreign Currency report should focus on countries with large overall trade and current account surpluses, not on countries with large bilateral surpluses. A country with a current account deficit (like India) should never appear on a Treasury watch list. Yes, that means less of a focus on China in the foreign currency report right now—China currently is a trade policy problem, not a currency problem. 2. The report should look closely for evidence that countries with a large current account surplus are intervening, directly and indirectly, to help keep their currencies weak. That means doing some financial forensics in those cases where existing disclosure is incomplete. Taiwan’s long-standing argument that it doesn’t hide anything by failing to disclose its forward position shouldn’t cut it. Swaps—exchanging foreign currency for domestic currency—can move foreign exchange off the central bank’s formal balance sheet, and we more or less know from the disclosed hedges of the Taiwan’s life insurance’s sector that it has a large domestic swaps counterparty. It also would require that the Treasury look more closely at the actions of government run pension funds and sovereign wealth funds, searching for what might be called “shadow” intervention. Singapore, for example, has held down its formal reserve growth by shifting reserves over to its sovereign wealth fund (the new transfer in May isn’t the first). Korea’s decisions to limit the hedging of its national pension fund structurally helped take pressure off the Bank of Korea to intervene, it should have received a lot more scrutiny from the Treasury than it did.* 3. A designation in the Foreign Currency report should serve as a warning that the United States could engage in counter-intervention against the designated country—as proposed by Bergsten and Gagnon. Counter-intervention is the most elegant sanction for “manipulation” (excessive intervention in the foreign exchange market). And, well, it wouldn’t be subject to legal challenge either—America’s trading partners have never been willing to negotiate away their authority to intervene in the market in a trade deal, and the United States equally has no legal limits on its own intervention either. The other potential sanction for manipulation—trying to offset the effect of an under-valuation through countervailing duties—at best only imperfectly fits into the existing trade rules.**  The last change is, of course, the most consequential. The United States, in my view, already has the legal authority to use the Exchange Stabilization Fund for counter-intervention. But actually doing so would be a significant shift in policy. In the past, the United States has typically intervened jointly with other countries, in a combined effort to signal that the market had overshot. Intervening to try to offset, rather than to complement, the intervention of another country is thus is about as far from the past use of intervention as is possible. Of course, the Exchange Stabilization Fund doesn’t have unlimited resources. But it is big enough relative to the countries that would most likely be caught in the initial cross-fire. That’s the advantage of introducing new policy when the world’s largest economies aren’t really intervening to hold their currencies down. And if the United States ever were to be at risk of running out of (counter) intervention capacity, an administration could approach Congress for the borrowing authority needed to raise a bigger stockpile of funds for counter-intervention (e.g. exempt such borrowing from the debt limit, up to some defined level). The idea, of course, would be to credibly signal to a country that was engaging in excessive intervention that its actions would be subject to counter-intervention, so it would adjust its policies in advance (e.g. either bring its current account surplus down, or reduce its intervention, or both. Or negotiate a path with the United States for doing so over time).   That said the practical consequences of changing policy along the lines I suggest would be very modest right now. To be sure, the dollar is currently quite strong. That’s clear in the trade data: U.S. manufacturing exports haven’t really grown since the dollar appreciated in 2014 (and service exports haven’t done much better).   But the dollar is currently strong because U.S. interest rates are (comparatively) high and that is pulling yield-seeking investors into the U.S. market, not because of massive intervention by America’s main trading partners.   And U.S. rates are higher than rates in many U.S. trading partners because U.S. fiscal policy is substantially looser than the fiscal policies of most of the United States major trading partners (China is the exception here, it too has a relatively loose fiscal policy and largely as a result it doesn’t have a large current account surplus).    Intervention is an issue when there is market pressure on the dollar to weaken, and other countries choose to counter-act that pressure because they don’t want a stronger currency to cut into their exports. And I do think such intervention damages the U.S. economy over time, and thus it makes sense to shift policy ahead of a change in the dollar’s path. For example, the U.S. recovery from the 2001 tech slump would have been substantially stronger—and much more robust and resilient—if it had been based on exports rather than a housing bubble. The large rise in intervention that started in 2003 thus did have important consequences. And similarly the U.S. recovery from the global crisis would have been stronger if it had been helped along by stronger U.S. exports in the years immediately after the crisis. Yet a number of countries, China included, intervened heavily in the four years after the global financial crisis to keep their currencies from appreciating back when the United States was far from full employment and policy rates were at zero (and fiscal policy was, politically at least, frozen and moving in the wrong direction from 2010 on).    The United States got a contribution from net exports to its growth in 2006 and especially in 2007 (and mechanically, the fall in imports in 2008 helped cushion the blow of the sharp fall in U.S. demand). But in part because of intervention outside the United States, net exports didn’t contribute to the U.S. recovery from 2009 on. The United States’ recovery was weaker as a result … A couple of smaller points here: Monetary easing through balance sheet expansion—the purchase of domestic financial assets—obviously has an impact on the exchange rate.*** But balance sheet expansion through the purchase of domestic financial assets (QE) is conceptionally distinct from balance sheet expansion through the purchase of foreign assets (“intervention”).****  Many countries with current account surpluses could do more to support their own domestic demand. The most pernicious policy mix is one that combines tight fiscal policy with heavy intervention to maintain a weak currency and strong exports. Korea’s post crisis policy (tight fiscal policy and intervention to block the won’s appreciation and keep the won at levels that helped Korea’s exports) should have received substantially more global criticism than it received at the time. In such countries, less intervention need not mean less growth—just a different kind of growth, as they have substantial policy space to support domestic demand. Countries with current account deficits generally should be building up their reserves as a buffer against swings in capital flows. Concerns about excessive reserve buildup should only arise when a country has a significant and sustained current account surplus. In my recommended policy framework, countries with current account deficits like Argentina and Turkey would have been free to build up large reserve buffers during periods of strong inflows without criticism from the United States. The bigger issue though is whether the costs associated with larger trade deficits—than would otherwise be the case in bad states of the world—warrant a shift in policy by the United States. A shift that would, at least initially, create additional sources of economic friction, including friction with countries that are now allies of the United States. Count me with C. Fred Bergsten and Joe Gagnon as among those who think the costs to the United States from excessive intervention by America’s trade partners are big enough over time to warrant a different policy.   * The report’s focus on bilateral imbalances led to the inclusion of Ireland and Italy on the Treasury’s watch list even though neither is intervening to weaken the euro. While Taiwan—a country with an enormous surplus, limited disclosure, and a history of intervention—has dropped out of the report.  ** To potentially fit with the WTO, counter-vailing duties need to be brought to offset sector injuries from a subsidy that provides a material financial contribution to production in another countries. That makes the sanction contingent on a bunch of industry specific legal cases—an across the board tariff in response to excessive intervention would be a more powerful sanction, but it would almost certainly not pass WTO muster. Basically, trade law wasn’t designed to allow currency sanctions; the fit is awkward. *** I have a table, prepared with help of Dylan Yalbir, that provides a guide to who would have met the current account surplus (of above 3 percent, I am not convinced that the recent move to 2 percent is warranted) plus heavy intervention definition of manipulation in the past. China obviously met it (and then some) prior to the global financial crisis. **** It is conceptually possible to purchase foreign currency without easing domestic financial conditions—purchasing foreign currency expands the domestic monetary base, but “sterilization” (raising reserve requirements, issuing domestic monetary bills and the like) allows the central bank to offset the domestic monetary impact (at least in principle) of its expanded external balance sheet.