MEASURING THE USD PREMIUM
How big is the Chinese economy? On one level, that question is easily answered – last year, China’s GDP was RMB 91 trillion.
For comparative purposes, though, what’s that worth in dollars? Authoritative sources will tell you that China’s dollar GDP in 2020 was $14.7tn. Those same sources will also inform you that it was $24.1tn. That’s a huge difference. On the first basis, the Chinese economy remains 30% smaller than that of the United States ($20.9tn). On the second, it’s already 15% bigger.
The explanation for this very big difference lies, of course, in the two ways in which economic data from countries other than the United States can be converted into dollars. One of these is to apply average market exchange rates for the period in question. For convenience, we can call this market conversion.
The alternative is PPP, meaning “purchasing power parity”. To apply this conversion, statisticians compare the prices of the same products and services in different countries. (One such common product is a hamburger, which is why, in its early days, PPP was sometimes called “the hamburger standard”).
The differences between market and PPP calibrations of GDP are enormous. Last year, world GDP was $85tn on the market convention, but $132tn in PPP terms. At the same time, the use of PPP conversion diminishes America’s share of the global economy. Last year, the United States accounted for 25% of global GDP in market terms, but only 16% on the PPP basis.
Using PPP instead of market conversion doesn’t make the economy ‘bigger’, of course. It just means that a higher dollar value is ascribed to economic activity outside the United States.
There’s no ‘right’ or ‘wrong’ way of converting non-American economic numbers into dollars. To a certain extent, it’s case of selecting the convention best suited to the topic being examined. Market conversion is appropriate for transaction values, such as trade, and cross-border assets and liabilities. PPP provides a better measure of the comparative sizes of economies around the world. (The SEEDS economic model produces parallel output on both conventions, though with a preference for PPP).
For macroeconomic purposes, the PPP convention is arguably more meaningful than market conversion, because it better reflects the economic scale of countries like China and India. Additionally, it leaves both market sentiment and short-term vicissitudes out of the process. PPP conversion has been with us for decades, and is carried out by reputable authorities, such that we can accept it as a valid and consistent alternative basis of currency comparison.
Market rates are determined by many factors other than economic comparison. FX market players have multifarious reasons for liking or disliking various currencies. Their opinions do not constitute economic ‘facts’.
An obvious example here is the reaction to the “Brexit” vote. British citizens obviously didn’t wake up 20% poorer on the morning after the referendum, but that’s what market dollar valuation of the UK economy implied. By the same token, market-rate conversion asks us to believe that the economy of resource-rich Russia is a lot smaller (at $1.4tn) than that of Italy ($1.9tn).
People in Russia, China, India and elsewhere are not poorer because FX markets don’t, relatively speaking, like their currencies. Currency undervaluation against PPP equivalence does make these countries’ imports more expensive, but it also gives their exports a competitive advantage.
Benchmarking the market dollar premium
For present purposes, the importance of having two FX conversion conventions is that it enables us to benchmark the dollar itself. Using world economic data going back over four decades, we can examine the relationship between the PPP and the market valuations of the dollar.
In 2020, for example, the GDP of the world outside the United States (WOUSA) was $63tn on the market basis, but $111tn in PPP terms. From this we can infer, either that market conversion undervalues the WOUSA economy, or that the market dollar trades at a premium to its PPP equivalent.
For convenience, we can call this difference the market dollar premium, and calculate the ratio for 2020 at 1.74:1. Put another way, the market dollar commanded a 74% premium over the PPP dollar last year.
There’s nothing abnormal about the dollar enjoying a valuation premium over other currencies. The dollar’s pre-eminence can be traced back to 1945, when America accounted for half of the global economy, and was the world’s biggest creditor. The dollar, after all, is the world’s reserve currency, and the benchmark against which other currencies are measured. Most oil trade continues to take place in dollars, providing a ‘petro-prop’ for the USD, because anyone wanting to purchase oil must first buy dollars.
This being so, it’s no surprise that PPP comparison reveals a market dollar premium.
What’s interesting, though, is the upwards trend in this premium.
In 1980, it stood at 30%. It reached 40% in 2001, and 50% during 2005-06. The market dollar premium reached 60% in 2009, and 70% in 2015. Based on consensus projections, the premium is expected to carry on rising, from 74% last year to 79% by 2026. Perhaps most strikingly, the dollar premium is twice as big now (74%) as it was in 1999 (37%).
Does the market’s attachment of a widening premium to the dollar make economic sense? It’s at least arguable that it doesn’t. Quite aside from the rise of economies such as China – and America’s falling share of world GDP – there are reasons to suppose that the economic pre-eminence of the United States is eroding, and that the market dollar premium, far from widening, should be contracting.
The most obvious negatives for the market dollar premium are to be found in the fiscal and monetary spheres. Starting in 2008, the Fed has operated monetization policies on a gargantuan scale, lifting the Fed’s assets from $0.8tn in June 2008 to $8.1tn today. Interest rates have been below any realistic estimate of inflation since the 2008-09 global financial crisis (GFC) and, with inflation now rising, are negative to the tune of at least 4.0%, and probably more. With the administration seemingly addicted to fiscal stimulus, and with the Fed apparently willing to go on monetizing deficits, these trends seem set to continue.
Scaling back or reversing QE – or, for that matter, raising rates to head off inflation – would prompt a greatly-amplified repeat of the 2013 “taper tantrum”, and tightening monetary policy could harm the US economy, would trigger sharp falls in asset prices, and would push up the cost of government borrowing. Neither monetization, large scale money creation or negative real rates can be considered positive for the value of a currency.
There’s a clear danger, then, that the US could push the dollar’s “exorbitant privilege” too far.
Meanwhile, the Fed also has to be mindful of the shadow banking system, sometimes called “non-bank financial intermediation”. This isn’t the place for a detailed consideration of shadow banking, but the system resembles an inverted pyramid, with very large assets (which have been put at $200tn) resting on a narrow base of collateral. Government bonds in general, and American bonds in particular, play a central role in this collateral.
Simply stated, a battle royal is likely to be waged between not-so-“transitory” inflation, on the one hand, and, on the other, pressing reasons for not raising the cost of money.
This might not matter all that much if the market dollar premium hadn’t risen as far as it has. The use of PPP for benchmarking isn’t common practice, but the calculations required for calibrating the market dollar premium aren’t exactly rocket-science – and the implications of this calculation are stark.
The conclusion seems to be that the dollar now trades at a more-than-exorbitant premium to other currencies – just as America is getting mired in a tug-of-war between stimulus and inflation.