Fed Governor Lael Brainard’s speech on central bank coordination last week was quite interesting—I think it should be read for far more than a signal on when the Fed is likely to next raise the policy rate.
For one, Brainard argues that the ECB (and BoJ’s) asset purchases have had an impact on global yields. Makes sense. The ECB and BoJ are both buying more than their respective governments are issuing, so they are reducing the net supply of eurozone and Japanese government bonds on the market. That forces bond investors into other assets—be it short-term deposits at the ECB, bank bonds in Europe, or U.S. bonds of various stripes.
That though wasn’t the central banking orthodoxy a few years ago. The spillover of U.S. asset purchases onto say European government bond yields was not apparent back when the U.S. was doing QE. In fact, QE2 generally coincided with generally rising eurozone government bond yields. In part because the eurozone was experiencing its own version of a self-created government funding crisis, as the creation of the euro meant that countries that previously issued bonds in their own currency were now issuing bonds in the ECB’s currency so to speak. And in part because QE2 coincided with a large U.S. fiscal deficit—it reduced the new supply of Treasuries investors needed to absorb, but it didn’t on net remove supply from the market.*
But the really interesting bit isn’t the technical argument about global spillovers from asset purchases. It is the hint—at least in my reading—that the Fed and the ECB should pursue different tightening strategies.
Consider a simplified global economy that constitutes three blocks. Two blocks have independent monetary policies and let their currencies float, and both have two central bank policy instruments—the policy rate, and the balance sheet. And the third block pegs, more or less, to one of the other blocks.
The block that is now in a tightening cycle has a current account deficit of around 3% of GDP (and a sizeable net external debt position, so an underlying stock imbalance too). The block in an easing cycle (for now) has a current account surplus of around 3% of its GDP (a bit more actually). The block that pegs has a current account surplus of around 3% of GDP (after adjusting for some, umm, irregularities in its trade data) and pegs to the currency of the deficit country.
The three blocks are obviously the U.S., the eurozone, and China. This model leaves a lot out. Japan and the newly-industrialized-economies (NIEs) combined have a current account surplus of well over 5% of their combined GDP. And the U.S. NAFTA partners all have sizeable current account deficits too.
Brainard postulates that tightening through increasing the policy rate has a bigger impact on the exchange rate than tightening through balance sheet reduction. And thus the choice of central bank policy instrument can have an impact on net exports, and ultimately the equilibrium current account deficit.
If I read her comments correctly, this argues for putting a priority in the U.S. on balance sheet reduction rather than raising the policy rate.
That is because the dollar is already strong, and already distorting (in my view) the composition of U.S. output.I cannot find evidence to support McKinsey’s argument that a combination of robots (automation), cheap natural gas, and rising wages in emerging markets are going to reduce the U.S. trade deficit in manufactures at current levels of the dollar. Rather the contrary.** At current levels of the dollar, the U.S. trade deficit in sophisticated “capital goods” is actually rising.
Basically, if given a choice, a country with a large existing trade deficit should choose to tighten its monetary policy in the way that puts the least pressure on the dollar, and the least pressure on the tradables sector.
Interestingly, the opposite holds true for the central bank of a large surplus region. It should choose to tighten through raising the policy rate rather than through balance sheet reduction. That would help bring the economy into better external balance.
It, in Europe, also probably has some positive financial stability spillovers. The nightmare scenario that reverses the eurozone’s current positive momentum is a blowout in Italian yields (see The General Theorist). And that risk would rise if the ECB is selling its Italian government bond (BTP) portfolio at the same time as regulatory efforts at “risk reduction” force the Italian banks to diversify their own government bond portfolio. ECB balance sheet expansion has expanded the supply of “safe” euro area assets while taking both duration and sovereign credit risk out of the market.
An ECB that tightens through rates and a Fed that relies on balance sheet roll off, in theory, would work together to in effect weaken the dollar and reduce the U.S. trade deficit and European surplus.
That at least is how I read this paragraph in Brainard’s speech:
“Let's turn to the case in which the two central banks choose to rely on different policy tools. In this case, one country responds to the positive shock by hiking its policy rate to reduce output to its initial level, while the second country responds by shrinking its balance sheet. The country that relies on the policy rate to make the adjustment experiences an appreciation in the exchange rate, a deterioration in net exports and some expansion of domestic demand, while the country that chooses to rely solely on the balance sheet for tightening experiences a depreciation of its exchange rate and an increase in net exports. Thus, while both countries achieve their domestic stabilization objectives, whether the requisite policy tightening occurs through increases in policy rates or reductions in the balance sheet matters for the composition of demand, the external balance, and the exchange rate.”
A weaker dollar would also make it much easier for the third block, China, to avoid a big depreciation that would raise its surplus. China now seems to manage against the dollar, more or less (though officially it manages with reference to a basket). It appears to have successfully carried out a modest controlled depreciation after the dollar’s 2014 rise—though the process itself wasn’t totally smooth. The yuan is now stable, in part because stability (against the dollar) creates expectations of stability and thus reduces outflow pressures, and contributes to stability (in reserves). And in part because China reversed its premature financial account liberalization.
Indeed, if China decides it wants to manage with respect to a basket for real, the yuan should appreciate against a depreciating dollar. Which would help the PBOC convince the market (and more importantly Chinese residents) that the yuan isn’t a one way bet, and also help keep the U.S. trade deficit from rising—China’s export volume growth in the second quarter was extremely strong, so I suspect China’s surplus is now poised to expand if the yuan remains constant.
It is potentially win-win-win, so to speak.
Of course, it all depends on the assumption that raising the policy rate and balance sheet reduction (quantitative tightening) can be calibrated to achieve the same level of domestic tightening with a different exchange rate impact. A two central bank, two instrument world has to differ in some fundamental ways from a two central bank, one instrument world in order to create new possibilities for de facto coordination.
And it depends on the assumption that the eurozone’s current momentum will allow the ECB first to scale back its easing and then start a tightening cycle.
It makes sense to me, though. And I think I see hints of it all in Governor Brainard’s speech.
* QE2 also didn’t weaken the dollar much, which led some to underestimate the foreign currency impact of ECB easing. The absence of a bigger impact on the dollar reflected two things I think: (a) the dollar was fairly weak at the time; and (b) foreign central banks—setting the ECB aside—intervened heavily to keep their currencies from appreciating. All this matters—net exports never contributed much to the U.S. post-crisis recovery (a rise in exports did help the US in 2007 and the first part of 2008, and a fall in imports helped cushion the demand blow of the crisis in 2008 and 2009, but net exports subsequently didn’t do much—until the dollar’s 2014 appreciation).
** I liked a lot of the recommendations in the McKinsey study, but it really seemed to suffer from an omitted variable—namely the value of the dollar. I guess that is too obvious to generate consulting fees. But it clearly matters. Technology (see Richard Baldwin) isn’t confined to a single country’s workers any more. And the postulated positive impact on cheap gas on U.S. manufacturing has clearly been trumped by other variables (one example suffices: Aluminum was one of the postulated winners from cheap gas as it is hugely energy intensive), and, well, the gap between U.S. and global gas prices has shrunk significantly since 2014 as global prices have come down. The reality is that the U.S. manufacturing deficit soared back in 2014 and 2015, and shows no signs of coming down (the rise in the deficit actually preceded the rise in the dollar, as there was a surge in imports in 2014—but the dollar clearly had the expected impact on exports).