As dollar-denominated liabilities grows beyond all reasonable means of repayment, only two ways out of the debt chasm remain: a hyperinflation in which nominal amounts owed are repaid in worthless dollars, or a debt-liquidating deflation resulting from a cascading domino chain of personal, corporate, and municipal defaults. As analysts struggle to understand which forces will prevail, the inflation-deflation debate of the 70s has been resuscitated. Recently, some deflationists have put forth an argument that a creature known as a “synthetic dollar short” will drive a substantial appreciation of the dollar.

Market sage Richard Russell writes:

Do you remember a few weeks ago I made a strange statement? I said that the US's huge mountain of debt amounts to a "synthetic short position" against the dollar. What does that really mean? It means that to pay off debt you need dollars.

In A Day Late and a Dollar Short by George Paulos and Sol Paha say:

Dollar debt is functionally similar to a dollar short position. Those who have borrowed money to exchange for another asset with the belief that the other asset will appreciate in dollar value have taken out the equivalent of a short sale of the dollar. Massive short sales have characteristics and consequences in markets and these characteristics follow patterns. If the US dollar follows these same patterns, then there is a crisis dead ahead. This will not be a crisis of dollar collapse but one of dollar scarcity. Few investors are prepared for such an event. The vast majority of investors are positioned to profit from a declining dollar and would be devastated if the reverse occurs. This analysis flies directly in the face of conventional wisdom and the best efforts of our monetary authorities to devalue the currency. A dollar debt crisis would be a classic deflationary event, but unlike other more gradual deflationary scenarios, we envision the possibility of a rapid rise in effective dollar value that is similar to the characteristics of a short squeeze. A rapid increase in the value of the dollar is something that few are prepared for and therefore would hurt a large number of people.

Similar is Rick Ackerman’s Goldbugs and Buffett Face Major Dilemma:

As I explained here recently, most of the world’s hundreds of trillions of dollars of debt is denominated in dollars, and this debt represents, implicitly, a massive short-position against the dollar. As such, all borrowers of dollars should be praying for inflation, since it would allow them to pay back what they owe in cheapened money. They could also pray for the one other thing that might do the trick – a dramatic rise in incomes over and above the rate of inflation.  Miracles do happen.

Paulos and Paha do an excellent job in their article of explaining the mechanics of a short sale, and in what respects debt and short selling are similar and different. Anyone not familiar with the process of short selling and how a “short squeeze” works should review their article before proceeding.

There are two key points in their article with which I agree. One is that in a fractional reserve banking system, money is loaned into existence as commercial bank credit, and that money is destroyed when the loan is paid down or defaulted. Widespread defaults would then cause a banking, or liquidity, crisis. The other point of agreement is that the US government owes a considerable amount of the total debt that they can always pay off by printing money. I will consider these two points in turn, and then present reasons that I think their analysis is insufficient to show that there is a large “dollar short” position.

Under fractional reserve banking, new money and new debt are created by the same stroke when banks loan money into existence.  When a loan is paid down or defaulted, the money supply contracts. Because loans are issued by one bank and deposited in other banks, the assets of one bank constitute liabilities of another bank.[1]

The creation of credit money constitutes a pyramid upon a base of central bank reserves. The larger the pyramid, the greater the danger that defaults in one part of the system will trigger a contagion of cascading cross-defaults that will topple the bankrupt the entire system. The inter-relationship between bank assets and bank liabilities is the mechanism by which debt defaults in one banking sector to cascade through out the entire system. In a crisis, as more banks default, more inter-bank liabilities are defaulted and more credit money vanishes.

Such a collapse contract the money supply, possibly to the level of central bank reserves.[2] he principle of supply and demand – that (all other things equal) a decrease in the supply of something increases its price – shows that the purchasing power of the money increases during a monetary contraction. According to this line of thinking, those who had either paper currency or some near-cash asset that survived the default would find their purchasing power in the domestic economy increased, while those with deposits in failed banks or who owned assets of bankrupt companies would find theirs wiped out.

Money ultimately has purchasing power because of the existence of things that it can buy. A definition of deflation is a decrease in the supply of money in relation to the supply of goods and services, resulting in an increase in its purchasing power. To understand this, imagine if the money supply was doubled but no more goods were produced: then the purchasing power of money would be about half.  The influence on the purchasing power of money would be identical if the production of goods suddenly dropped by half while the money supply remained about the same.

The “dollar short” scenario of an increasing value of the dollar assumes that the supply of goods remains approximately unaffected by the liquidity crisis. But if both money and goods are contracting, the purchasing power of money can only increase if increase the money supply contracts faster than the supply of goods.

To forecast how the purchasing power of money might change during a banking collapse, it is important to take into account that banking system failures are not isolated from the rest of the economy. Typically a banking crisis and a production crisis (known as a recession or depression) come about at the same time from the same cause: the distortions and mal-investments in the economy created during the previous credit-driven boom. The production crises will most certainly be exacerbated by the banking crisis as the lack of solvent banks makes commerce in general more difficult.

The banking crisis does not cause the recession. In fact it is the beginning of the recession, as business fail to find buyers for their goods at the prices they had anticipated, or a rise in inflation squeezes profit margins, that kicks off the banking contraction.[3]

An economy experiencing a recession will see a reduction in the total the supply of things that money can buy. This includes not only goods and services, but financial assets, representing claims on goods. The supply of financial assets will contract as bonds are defaulted and public companies go bankrupt, leaving their stock holders with nothing. In our recent history, where the credit-driven boom of the 90s was manifested in a stock market mania, the supply of financial claims typically multiplied as expanding liquidity chased after financial assets. The net change in the purchasing power of money in a bust will depend on whether the supply of money proper or the supply of things that money can buy contracts faster.

The onset, or even the threat, of a deflationary collapse would surely be resisted by authorities with an inflationary policy response.   Deflationists emphasize that component of money creation occurring within the commercial banking system. Noting that this money creation is accompanied by a corresponding debt creation, they argue correctly that this process is inherently self-limiting due to the ever-increasing burden of debt service payments. The requirement to make interest payments on existing debt eventually limits the ability of borrowers to take on more debt.

Deflationists argue from this, that debt deflation will always trump inflation. The problem with this argument is that it proves too much. If this were universally true, then how could there ever be a hyperinflation? Debt would choke it off before it got started. And yet there have been many historical instances of hyperinflation.

There cannot be a short squeeze for something of which the supply is not limited. These analysts ignore the other money creation channel that a central bank can apply: monetizing debt directly The Fed currently is allowed to purchase Treasury debt (either newly issued or on the Treasury debt market) in its so-called “open market operations.” To do so, the Fed writes a check to the Treasury, creating out of nothing the money to purchase the bond. The seller could be the US Treasury or a bond holder who had purchased the bond some time before.

The monetization of any asset is similar to the purchase of Treasuries: the Fed buys something and pays the money into existence with a check. Asset monetization has the potential to arrest or head off the debt-default process. As new money is created and deposited in banks, it increases the equity of the banking system, alleviating the deflationary pressures that otherwise would force banks to contract their loan portfolios. 

In the pyramid scheme known as central banking, commercial banks have demand deposit accounts with the Fed, much in the same manner as bank depositors have them at the commercial banks.[4] The seller of a bond deposits their check into their account in a commercial bank.  The bank’s account with the Fed is then credited by that amount. To the extent that the Fed is able to monetize assets it can increase bank reserves, upon which banks can increase the money supply by pyramiding further.

In a crisis, the Fed, or agencies set up to act on its behalf  (perhaps public, perhaps in secret) with its checkbook in their back pocket, could purchase assets through the capital markets, such as the stock or bond exchanges, or could buy assets directly off the books of the banks.

This strategy is beyond hypothetical. The Fed has issued a number of position papers and studies over the last few years exploring “unconventional measures” that a central bank could take to stave off a deflationary collapse. Fed Governor Ben “Helicopter” Bernanke has mused publicly in a series of speeches along similar lines, in which the Fed’s “printing press” is prominently mentioned. The measures under consideration amount to the direct monetization of various assets. In their discussions, monetization is not limited to paper assets, but curiously includes the mention of gold mines, among other things. 

Currently there are limits to what kind of assets the Fed is allowed to purchase by its charter, and legal issues about the constitutionality of the Fed going beyond these limits. However, to guess which way this question would be settled, we cite a tenet of the inflationist view that in a social democracy, no politician or government agency will voluntarily embark upon a path toward the financial ruination of the majority of voters if inflation is an option.

Facing a crisis the Fed do what central banks always do: print more money. Constitutional restrictions and regulations are always ignored or over-ruled when a crisis unfolds. Whether the Fed could succeed in averting a deflationary collapse with a hyper-inflation is not certain. But who would question that unconventional measures would be tried? 

In case you have any doubt, consider the words of Fed Governor Bob “SUV” [5] McTeer:

In the early '30s when that episode started, there were a lot of bank failures that wiped out a lot of money and I don't know what the Fed could've done under those institutional arrangements but it, certainly, wasn't in there pumping out new money like we would be doing today. Today, every time we have a major emergency, you know, the first thing we do is get on the microphone and open up - you open up a spigot. I mean look at what happened in 9/11. I mean on 9/11, we just flooded the markets with liquidity because of all the damage in New York, you know, all these New York banks and investment banks, they're receiving billions in payments every day and they're making billions in payments and you know, if they don't receive it they can't make it and so, just a hitch or two in that system can bring the thing down. Well, we just pumped enormous amounts of liquidity in there through open market operations and our check clearing system, which the Houston branch is very involved in, we decided to give credit for checks deposited with us, on the next day when it would normally be done, even though all the planes were on the ground. We couldn't collect the checks but we pretended we were collecting the checks and we gave credit for those checks, created enormous amount of float, which by law, we're supposed to treat as a real cost, to us, but since we're more a public institution than a private institution, we decided not to put our cost situation ahead of the public good, anyway - I'm getting too far off. We know how to handle those things better now, not well enough but not bad.

There is a historical example that illuminates the likely response of a central bank to a debt default crisis. During the Great Depression, a large number of mortgage borrowers were under water and in danger of losing their homes to the banks, who would have then been saddled with a portfolio of homes worth much less than the debt that financed them. Home prices had appreciated during the speculative excesses of the 1920s. To avoid wide spread home loan defaults, the precursors of the modern GSEs (Fannie Mae, Freddie Mac, and the Federal Home Loan Bank) were set up to purchase the impaired mortgages from the banking system and renegotiate the terms of the debt to avoid default. At the time, the Fed was more restricted than it is today in its ability to create unlimited amounts of money out of nothing because the United States was still on a gold standard.

A second facet of this problem that I believe is missing from the “dollar short” analysis is the role of foreign holdings of US debt. If the US were a closed economy with no foreign exchange, a domestic scramble for liquidity would work itself out along the lines analyzed above, possibly resulting in an increase in the domestic purchasing power of the dollar.

But the US is not a closed economy – there is an enormous flow of both goods and financial assets between America and the rest of the world.  It is not clear to me from the writings cited above whether the “dollar short” deflationists are viewing the issue as if the US were a closed economy experiencing a monetary contraction, or if they are forecasting an increase in the foreign exchange value of the dollar.

The foreign exchange value a currency can rise for two reasons: a decrease in supply of that currency, or a net increase in foreign demand by holders of other currencies. Only buyers who would exchange either goods or other currencies for dollars can create a net increase in foreign demand for dollars.

Forecasting how a collapse of the US banking system would affect the foreign exchange value of the dollar is difficult. All things being equal, a decrease in the total global supply of dollars (while other national currency supplies remained the same) would result in an increase in the exchange value of the dollar against other currencies. But all things are not equal for several reasons. One is the contraction on the goods side from the recession, described above. The other is that a banking crisis in the US would surely affect the foreign demand for US dollars in other ways that will be explained more fully below.

Richard Duncan, in his book, The Dollar Crisis (see my review), tells the story of how the ability of Americans to fund their consumption by the issuance of debt came about as a disastrous consequence of the failed Bretton Woods monetary system. In this system, the world was placed on a dollar standard, while the US dollar alone remained on a gold standard. When the US defaulted on its obligation to redeem dollars for a fixed amount of specie when Nixon “closed the gold window”, the rest of the world’s central banks were left holding vast and rapidly growing dollar reserves.

The buildup of dollar-denominated debt relies on the willingness of foreigners, mostly foreign central banks, to fund an increasing amount of US indebtedness each year. Overlooked by the deflationist analysis is the huge supply overhang of US dollars that constantly threaten a collapse in its foreign exchange value. The dollar is vulnerable should the largest holders, who already own too many of them, decide to sell. What keeps them from doing so is (at least in part) that doing so would spark a mad rush for the exits that would destroy the value of their own remaining reserves.

The “dollar short squeeze” is mistaken in looking at debt as a short position in that indebtedness only represents a latent demand for dollars when borrowers have the ability to demand dollars in the future. In order to demand dollars, a buyer must supply something that can be exchanged for dollars. Debt that is not backed by the ability to supply something in exchange does not represent latent demand and cannot result in a short squeeze.

Most of this article is concerned with how a crisis might work out. But for a moment, let is consider how it would look if the US trade deficit were to be reversed by an organic demand for dollars. In that scenario, the deficit country (e.g. the US) would have to see the dollar fall far enough in relation to other currencies to cripple its ability to import, and to make its exports cheaper to the rest of the world. Americans would then, over a period of time, have to export goods that they had produced for foreign currency that they could exchange for dollars to buy back and extinguish the foreign debt. The relevant point here is that being indebted to foreigners is bad for the value of your currency.

For every other country that has had a banking crises, it has been accompanied by a collapsing currency (think: Argentina) and international bond defaults because the ability of the country to produce goods for export (with which to buy foreign exchange) is reduced as the recession or depression works its way through the country’s real (non-financial) economy.

An ironic feature of the international financial system in recent years is that a series of credit bubbles driven by the ever-expanding dollar system have resulted in a stronger dollar. As Duncan explains in his book, US dollars held by foreign central banks serve as part of the monetary base upon which those countries can expand their domestic money supply, driving a boom-bust cycle. A series of crises – in Mexico, Asia, Russia, Argentina – resulted the accumulation of even more dollar reserves in a rush for the “safe haven” currency. Whether this process has any limits will be a key to understanding the future of the dollar’s exchange value.

After decades on the world dollar system (and the recent financial crises), there is an estimated $500 billion in US currency (coins and notes) that circulates outside of the US. Currently, foreign holders own over 20% of US treasury debt, a figure in the trillions of dollars. If the largest foreign buyers of US dollar debt (who also are the largest holders of dollar-denominated assets) reduced their demand for dollars during a US liquidity crisis, or even became net sellers, then demands for dollars fall. 

Contrary to the “short squeeze” analysis, the supply of dollars being offered for sale might rise rather than fall as foreign buyers seek to unload their positions. So far we have been looking at US conditions only, but a crisis that originated in the US would put the entire world financial system under stress. The “scramble for liquidity” described by the “dollar short” theorists would take place in all countries.  The reserve position of US Treasuries would be one of the largest assets on their books. Whether they would be find buyers would be another question, since the Fed cannot print Yen or Euros.

The composition of dollar-denominated debt held as an international reserve asset consists mostly of US Treasuries and GSE-insured mortgage-backed securities. Treasury debt is the least likely of any asset to default because of printing press. Although Al Greenspan has gone to great pains to deny that the issues of Fannie and Freddie are government-guaranteed, he has always been there in time of need to turn on the confetti machine and bail out any entities deemed “too big to fail”.

With respect to the foreign exchange value of the dollar, most of the borrowers of dollars do not resemble short sellers, in that they have not borrowed dollars and sold them for non-dollar denominated assets that they could sell to buy back dollars. US households, corporations, and governments are not “short dollars” “functionally”, “synthetically”, or otherwise. They have borrowed the money from foreign central banks and “sold” it for consumption goods of one form or another.  

Consumption goods cannot easily liquidated for currency. While it is true that the American consumer who has bought a DVD player from Japan with his credit card is not short dollars might enjoy a bit of inflation to lower the real cost of his debt, he is not short dollars and long yen. Let him try to sell his DVD player on eBay for a few Yen so he can buy back some dollars and “cover his short”.  The banks of China and Japan have obliged them by purchasing the dollars from their domestic producers, and then loaning those dollars back to the Federal Government through the purchase of Treasury debt, and now are very much long dollars.

Paha and Paulos write, “the vast majority of investors are positioned to profit from a declining dollar and would be devastated if the reverse occurs.” It might be true that the vast majority of gold bugs who visit financial-oriented web sites are positioned this way, but compared to the trillion dollar holdings of the Bank of Japan who are “net long” the dollar, this is not true in the foreign exchange case.

There are hedge funds who have borrowed dollars and who will have to sell other assets to buy back dollars at some point. But these funds might have purchased emerging market debt, junk bonds, or gold stocks, all asset classes that have seen a large sell-off in recent months. This sell off might have marked the unwinding of their positions in what might have been the “dollar short covering” rally predicted by the analysts.

For countries that do not have the right to print the reserve asset, the foreign exchange value of their currency depends ultimately on their ability to produce goods that can be purchased with that currency. But the dollar has benefited from extra foreign demand, above the level of dollars required to purchase American goods, due to its status as the international reserve asset. OPEC oil is one example of something that the rest of the world needs dollars to buy.

Would a recessionary collapse of the supply of goods available in the US affect the dollar’s privileged status? Would a banking crisis in the US threaten the privileged position of its currency? If demand for dollars as a reserve asset were to diminish, that can only be seen as most bearish for its exchange value. If dollar creditors finally recognized the impossibility of their debt every being repaid, would they start to sell their assets? Or would they sell Treasuries and start buying other riskier US assets, in an attempt to triage the losses by propping up the system a bit longer?

The inflation-deflation issue is indeed quite puzzling due to the myriad of interacting factors, both correlated and cantilevered. I do not mean to suggest that the inflationary response that engendered by a deflationary crisis is in any sense a sustainable solution to the excesses of the preceding boom. Such a program would surely result in a hyperinflationary recession, rather than a deflationary one. The Austrian view is that that resource mis-allocations of a credit-driven boom require a recession to be cleansed. This tells us that distortions in the real economy cannot be erased by changes in monetary policy (conventional or un-). 

As a policy recommendation, to allow the deflationary bust to do its work would be the best path. But as financial writer James Grant has written, if there is one thing that governments excel at, it is debasing their own currency. I would not bet on the sudden onset of incompetence.

[1] The process of banks creating money by pyramiding credit on top of their cash and the claims of other banks is explained very clearly in Murray Rothbard’s book The Mystery of Banking, see http://www.mises.org/mysteryofbanking/mysteryofbanking.pdf.
[2] This is also explained in Rothbard’s book.
[3] I am relying here on the Austrian theory of the business cycle pioneered by Ludwig von Mises, see http://www.mises.org/manipulation/manipulation.asp. 
[4] See Rothbard on this as well.
[5] McTeer said during the early days of the tech bust, that if everyone would go out and purchase an SUV, the economy would be fine.  For more examples of the wisdom of this great economic inflationist, see http://www.mises.org/blog/archives/002049.asp.

© 2004 Robert Blumen