The Dollar Has Peaked – Part II

In part I of this series, I claimed that the US dollar shares some characteristics with risky assets, notwithstanding its reputation for being a safe haven. In this installment I elaborate on this view, one that I first developed back in 2000 while questioning the validity of traditional foreign exchange forecasting models.

In the late 1990s the US economy was booming, helped along by the tech bubble and, in no small measure, by the Federal Reserve’s controversial decision to facilitate the bailout of Long-Term Capital Management in late 1998. At the time, the Fed sought to ensure sufficient liquidity was available for financial markets. But, in the event, the emergency liquidity that they provided seemed to leak out everywhere and led to fresh, arguably irrational exuberance in risky asset prices – to paraphrase former Fed chairman Alan Greenspan.

As the bubble deflated in early 2000, the Fed began cutting rates – and quite aggressively, with former Fed chairman Ben Bernanke, a governor at the time, warning of the dangers of outright deflation in 2002. Yet throughout this period, the dollar remained strong. Indeed, the dollar was overvalued when measured in purchasing-power-parity (PPP) terms and both nominal and real interest rate parity terms.

This was a headscratcher for those like myself who had predicted that US interest rates would decline sharply relative to those elsewhere and therefore that the dollar was likely to weaken as per the conventional foreign-exchange wisdom. And so I went looking for answers. One place I looked was at the changing nature of cross-border capital flows.

The Changing Composition of Cross-Border Capital Flows

By the late 1990s, the global economy had become far more financialised than had been the case in the previous major dollar downtrend beginning the mid-1980s. Indeed, entirely new asset markets had been created, notably those for large-cap ‘tech’ stocks, ‘junk’ corporate bonds, and risky emerging market debt. These new markets were almost exclusively dollar denominated. Any global investor wishing to gain exposure would need to acquire dollars in order to purchase these assets.

While foreign investors might hedge some of their exposure to these dollar-denominated assets back into their base currency, in practice few investors hedge 100%. Indeed, to hedge 100% of a risky asset is itself a risky proposition because it is impossible to know with any certainty what the asset is likely to be worth in nominal (much less real) terms in a month’s time (much less a year). And so with the growth of new, investible asset classes in the 1990s arose large speculative cross-border flows not merely into government bonds or bank deposits – the traditional places for such speculation – but into these new areas where risk premia dominate.

It was in 2000, while analysing this phenomenon, that I coined the term ‘speculative parity’ to challenge the conventional wisdom that some form of purchasing-power or interest-rate parity was the most robust model for forecasting future foreign exchange rates. The tail of asset speculation, as it were, was wagging the dog of more traditional low-risk asset flows and that of actual traded goods and services, the traditional drivers of foreign exchange rates in the classical (or neo-classical) economic world.

The seed for this idea of ‘speculative parity’ was an excellent piece of original research by Cameron Crise, featured in the Economist in 2000, focusing on the equity risk premium as an FX indicator. In my research and modelling work, I generalised this to include the risk premia for all major asset classes, including high yield corporate and emerging market debt.

Looking across G10 FX, applying multivariate regression analysis, I noted how some currencies, including the Swiss franc, still appeared to be dominated by interest rate parity, in particular when using real interest rates rather than nominal ones. The so-called commodity currencies, such as Aussie or Canadian dollars, still exhibited a strong correlation to industrial commodity prices. But the surprising result was that, in the case of the dollar itself, the spread of dollar high-yielding debt to Treasuries was the single most important explanatory factor, with the equity risk premium and dollar-denominated emerging market debt risk premia also statistically significant.

When grouped together – and statistically they should be treated as correlated variables in any multivariate regression – these risk premia showed that, as premia generally rose, the dollar tended to decline, and vice-versa, and with a lag of some 2-3 months.

And so the puzzle pieces began to fall into place: in 2000, the tech sector crashed, but initially the broad stock market held up reasonably well. It was only in 2001-02 that the broader stock market corrected meaningfully. Meanwhile, as the Fed slashed rates in 2001-02, in the markets for emerging market and high-yield corporate bonds, yields declined along with Treasuries, pushing up prices. Leaving aside the shock of the Sep11 attacks, spreads did widen modestly but less than implied by forwards, and returns were still generous, with Baa yields averaging over 7% yet CPI inflation declining to nearly 1% by 2003.

The Dollar Finally Peaked in 2002

It was only after risk premia generally had risen for some time and financial markets priced in that the Fed would take real interest rates to zero, in early 2002, that the dollar began its decline. Global investors were now not only de-risking and thereby selling dollar-denominated risk assets but, with US interest rates offering no meaningful spread to Europe or Japan, they were repatriating. Roughly two years later, in 2004, a general recovery in risky asset markets was underway. By that time, however, the dollar had fallen substantially. It would continue to drift lower until the global financial crisis arrived in 2007, but by then the bulk of the dollar’s adjustment was already completed.

The Dollar in the 2008 Global Financial Crisis

In 2008, as liquidity in most markets collapsed, there was a general rush into safe-haven assets, including US Treasury bonds and bank deposits. The dollar soared, briefly, before extreme, unconventional Fed policies, TARP bailouts, and other unprecedented programmes led to a pullback in 2009. Indeed, in real effective terms, the trade-weighted dollar would reach a new low in 2010, before entering the current, now decade-long uptrend.

While the US dollar therefore clearly demonstrated classic safe-haven characteristics for a few months in 2008, I believe that this sort of behavior is now the exception that proves the rule. When leveraged dollar positions must be liquidated, and the money supply implodes due to distress in money-market funds—which are only a marginal form of money in any case—then the Fed, in its role as lender of last resort, steps in to prevent it.

In other words, in the one set of conditions in which the dollar should indeed squeeze higher, the Fed won’t let it. We saw this again late last year when the Fed stepped into the repo market and, based on what the current chairman, Powell, said just yesterday, this sort of action, however unconventional the tools used, is now perceived as a core part of the Fed’s mandate. The dollar’s future value needs to reflect this, too. In my opinion, it currently doesn’t.

The Dollar Outlook

As asset markets have grown in size relative to GDPs, speculative flows have become ever more dominant, and the dollar remains the currency of choice for risk asset issuance and speculation. For this reason the relationship between risk premia generally and the dollar is arguably even more powerful now and helps to explain why the dollar should continue to behave as a risk asset under normal conditions.

There is, however, a more traditional reason for why the dollar’s fate may well remain linked to risk asset speculation: the chronic trade deficit. While the point of this series is to point out that cross-border flows in goods are now tiny when compared to speculative flows, the underlying fundamental economic reality cannot be denied when thinking longer-term. As a result of the large net foreign debt position already accumulated by the US, it now needs roughly $2bn a day in fresh foreign dollar demand to cover the current account gap and prevent the dollar from declining.

If the US is going to remain a chronic net debtor, as it now has been for some forty years, it will need to attract this foreign capital just to maintain a stable currency. If it is unwilling to offer a material interest rate differential, in real terms, that will be difficult unless risk appetite recovers. Therefore, if risk appetite fails to recover and the Powell Fed continues as it is already doing to prevent any form of general liquidity squeeze in the banking system, I think the dollar has a long way to decline.

The Dollar Has Much Potential Downside

Eventually, although this would be a rather bold forecast, the dollar could decline to the level implied by more fundamental measures such as purchasing-power parity. It could decline even further than that were the US to rebalance away from chronic imports towards export competitiveness. Economic history demonstrates that chronic trade imbalances are unsustainable. That which can’t go on forever, won’t.

One of the strongest arguments against the thinking above is that the dollar remains the dominant reserve currency, and so there will simply always be a bid for reserves. But this is, in a way, tautological thinking. Economic history also instructs that reserve currency status doesn’t last forever. Were the US dollar competing with other currencies on anything close to a level playing field – one in which there was no privileged reserve status – it is difficult to imagine that its core currency fundamentals (chronic twin deficits; a huge accumulated net foreign debt position; and demographic trends that show it closing the gap with its main trading partners) would make a good foundation from which to make a bullish case for the dollar longer-term.

These are structural issues to be sure, but as the Chicago School reminded us in the 20^th century, balance sheets matter. This is ultimately as true for public as for private balance sheets, as periodic emerging markets crises have showed us in recent decades. Yes a heavily indebted public sector, with the help of a central bank, can buy time, but time has a cost. And if the time is not used to rebalance the economy away from imports towards exports, then a balance of payments (BoP) crisis will eventually arise.

Could the US Face a Future Balance-of-Payments Crisis? The Risk is Not Zero

There is simply no way to print your way out of a BoP crisis. Ask Great Britain, which discovered this the hard way in the 1970s. In my opinion, if it could happen to Great Britain, it could happen to the US. Great Britain was in dire straights in the 1970s, running twin deficits of its own and facing growing social unrest due to widespread unemployment and other social issues. Sound familiar?

For those who share the view that the US dollar is at greater risk than is generally appreciated, I would recommend overweighting the so-called ‘surplus currencies’ instead. These are those that, in contrast, run chronic trade surpluses and are therefore net contributors rather than consumers of global capital. This would include both the euro and the yen.

For those concerned about the euro’s future, the yen emerges as the better choice. Yes, Japan has its own somewhat unique set of problems, including a huge public debt. But when considering the foreign exchange implications of this, note that Japan owes this debt to itself, not to foreigners. It is an internal imbalance only that external factors, e.g. yen devaluation or revaluation, cannot directly address.

Externally, notwithstanding energy import dependence, Japan has been mostly in balance over the past decade or so, and with a huge legacy net foreign credit position. The Japanese are, in modern economic history, the greatest savers there ever were. They may have demographic challenges, but they have saved and accumulated a huge amount of foreign assets to help fund their retirement, as it were.

There are those who will remain skeptical of both the euro and yen. That leaves only smaller currencies, some of which are also attractive in my opinion. But don’t forget gold. Gold is the only global money that does not have a national ‘balance sheet’ at all. Gold has no fundamentals as such. It just is. As economic history repeatedly demonstrates, when currency fundamentals are generally poor, this can be an advantage.

In closing, recall that the long gold bull market of the 2000s got going about two years’ prior to when the dollar started its general decline in 2002. Gold had been declining on trend for years and had fallen out of favour. It would be difficult to make the argument today that gold is out of favour, as it frequently makes the financial headlines. But as the analysis above shows, gold has good reason to be in favour, especially with those who share concerns about the fundamental health of all major global currencies.

John Butler has 25 years experience in international finance. He has served as a Managing Director for bulge-bracket investment banks on both sides of the Atlantic in research, strategy, asset allocation and product development roles, including at Deutsche Bank and Lehman Brothers.

(The commentary contained in the above article does not constitute an offer or a solicitation, or a recommendation to implement or liquidate an investment or to carry out any other transaction. It should not be used as a basis for any investment decision or other decision. Any investment decision should be based on appropriate professional advice specific to your needs.)