A Crisis Reinforcement Channel Wakes Up
We have recently discussed the alleged “dollar shortage” with Mish and several others. The first thing that springs to mind when hearing of a “dollar shortage” is this: why should there be a shortage of a fiat currency that not only can be printed at will, but the supply of which has in fact more than doubled since early 2008?
Photo credit: Christopher Furlong/Getty Images
Of course those who are in a sense short of dollars are the many dollar borrowers abroad, who need to service a $8.6 trillion debt load. Here is a chart published by the BIS we have shown previously:
Stock of dollar loans issued to non-banks abroad (lhs), annual growth rate (rhs).
To the extent that these dollar borrowers have income that is not denominated in dollars, they may indeed be in trouble now, as the dollar’s exchange rate has relentlessly soared. This incidentally slows down the accumulation of dollar reserves by central banks in countries with a mercantilistic economic policy (since they don’t feel the need to weaken their currencies anymore).
Over time, this should work as a counterweight to the dollar’s strength, as domestic money supply growth in these countries will tend to slow down as a result. In addition, other currencies are currently beginning to replace the US dollar for new borrowings. These days the euro and the yen have become popular funding currencies – initially, this tends to exacerbate the price trends that are already in place, but at some point, the problem currently faced by dollar borrowers will migrate to euro and /or yen borrowers.
What is quite interesting is where the majority of dollar borrowers are located. Most are actually domiciled in emerging markets, the currencies of which have been especially weak. So there undoubtedly is a problem, but as we will explain below, the problem shouldn’t be termed a “shortage”.
Most dollar borrowers outside of the US are located in emerging markets
In the discussion alluded to above, we made the following remarks (which were in part already published by Mish), to which we add a few additional thoughts below:
The visible currency effects (such as a soaring dollar exchange rate and funding gaps, see below) are usually mainly a consequence, not a cause of crisis conditions. Of course, the recent rise in the dollar was initially triggered by perceptions of monetary policies between the US and other currency areas diverging – everybody expects the Fed to hike rates, while rates are being lowered everywhere else. It should be added to this that there are of course feedback loops at work: the stronger the dollar becomes, the more difficult it will be for dollar debtors abroad to service their debt, so any future crisis situation will tend to feed on itself. Note in this context that if a debtor has hedged his dollar exposure, the associated currency risk has not disappeared – it has merely been shifted to his counterparty.
As can be seen from the charts above, borrowing in dollars has increasingly been effected via bond issuance in recent years – however, although their share of dollar loans abroad has decreased, bank loans extended to overseas borrowers have still continued to grow. If funding of dollar assets was a problem in 2008, it would be an even bigger problem today if a crisis that once again creates major distrust among financial intermediaries were to hit.
A Shortage of Dollars?
There are large amounts of dollars held in deposits abroad (mostly in Europe), and fractionally reserved banks are of course lending them out. They routinely have sizable asset/liability maturity mismatches (banks do lend out money held in demand deposits after all, which is the “ultimate” maturity mismatch), but this is true of banks everywhere, regardless of the currencies employed.
Prior to the central bank swap arrangements (this is in reference to the dollar swap windows the Fed has established with numerous central banks abroad), overseas based commercial banks could not turn to their own central banks for “lender of last resort” services with respect to funding of dollar assets when dollar deposits suddenly fled (US money market funds are e.g. a major source of these dollar deposits, and they pulled back when the US mortgage credit bubble blew up).
This concern is by now academic, since the swap lines now do exist, and are theoretically unlimited. Incidentally, US taxpayers have not suffered any harm, as all the swaps have been paid back (and even if they hadn’t been paid back yet: no-one was actually “taxed” to create them. The money was created from thin air, with the push of a button). The chart of central bank liquidity swaps below also supports our above contention: dollar funding problems in overseas banking systems are primarily a consequence, not a cause of crisis conditions.
As noted above, the largest amount of dollar deposits held abroad is located in Europe. European banks are therefore also the biggest lenders in dollar credit transactions between non-US residents. They are also the biggest lenders to emerging markets in general, so any problems there are bound to impair the situation of European banks. In fact, three quarters of all dollar-denominated lending to non-US resident borrowers is funded with dollars held abroad – only one quarter comes directly from US lenders.
We currently see euro basis swaps trading in negative territory again, which is indeed an indication that dollar funding conditions have tightened. Shorter term basis swaps have not deteriorated as much as the longer term ones – usually, when there is really a big problem, the shorter maturities will move into negative territory at almost the same pace (this is just an empirical observation). This may yet happen with a lag though.
Something is definitely stirring though – and it happens concurrently with the euro showing extraordinary weakness and in concert with on-and-off market concerns about the situation in Greece (although we would argue these concerns are more “off” than “on”, so they may only be playing a small role. It is of course a role that harbors some potential to become bigger).
As noted above, there are approximately $8.6 trillion in dollar-denominated debt outstanding abroad, a not inconsiderable amount. The basis swap “penalty” rate evident in euro basis swaps could be an indication that there is now a measure of distrust in the system: dollar lenders need to be recompensed for taking credit risk. The actuarial value of the fx basis should normally be at zero – if it isn’t, something is amiss. The main reason for the fx basis to drift into negative territory are usually concerns about the creditworthiness of banks.
Readers who want to learn more about short term swap markets can download a guide by CS (pdf) we have made available for such occasions. In addition to this, here is a recent document by JPM (pdf) that discusses the current signs of growing dollar funding difficulties (referred to as a “shortage” throughout, but we believe this is not the best choice of words. “Funding gap” is perhaps better). JPM argues that since no banking crisis is in evidence, the perceptions about diverging central bank policies must be the main reason for the developing supply/demand imbalance. We are not entirely sure about this though.
Per experience, euro basis swaps will decline deeply into negative territory before European banks will apply for funds from the Fed/ECB swap window, since the funds provided by central banks via these swaps don’t come for free either. In fact, they are relatively expensive if memory serves, which means that once one sees banks borrowing from this facility, the situation is probably quite dire already. Obviously, since the amount of central bank liquidity swaps currently outstanding is at zero, there is no evidence yet of an acute funding problem. What there is are however the first smoke signals that something of this sort could soon be afoot.
We think it is actually not quite correct to speak of a “dollar shortage” in this context. The domestic US dollar money supply has more than doubled since 2008 (in terms of TMS-2), and a sizable amount of dollar creation has “escaped” into accounts held abroad as well (read: the dollar money supply held outside of the US has also increased). It is not a shortage of dollars, but a growing unwillingness of dollar holders to lend them out that could create a problem for banks in need of dollar funding.
Bond Issuance vs. Bank Lending
Most of the new dollar lending to overseas borrowers has happened via bond issuance, and banks have shut down a lot of their proprietary bond trading due to new regulations and have cut back mightily on assets with a risk weighting higher than “zero” in order to comply with higher tier one capital requirements (this can sometimes backfire, as in the case of senior “guaranteed” Heta bonds). Given all this, should it not be assumed that any crisis in the context with dollar debts of overseas borrowers would be of – relatively speaking – little concern to banks?
The biggest losses would after all be suffered by investors in these bonds, most of whom are non-banks. However, this argument overlooks the interconnectedness of credit markets. Banks could find themselves affected indirectly. A borrower who defaults on a dollar-denominated bond and is forced into bankruptcy will also default on his other debts – and these will invariably include many bank loans. Many investors are in turn borrowing funds to leverage their investments in bonds. With interest rates at rock bottom levels everywhere, ample leverage is reportedly employed to juice returns.
Moreover, what is commonly referred to as the “shadow banking system” is also interconnected with banks. Losses in corporate bonds (which don’t necessarily have to stem from defaults – a rise in rates for whatever reason could already produce painful losses for leveraged investors and would set off a wave of forced selling) will reverberate across the financial system. Note that the fact that banks are no lonfer as active in corporate bonds as they once were is likely to greatly exacerbate volatility.
How things will precisely play out when the next financial crisis arrives is not knowable in advance. There are always surprises, and all we know with certainty at this point is that the next crisis will look different from the last one. For instance, even during the mortgage credit crisis several flashpoints of the crisis were undoubtedly a surprise for most market participants (think e.g. about AIG’s credit default swap portfolio blowing up – no-one expected this in advance to our knowledge).
There are some indications that trouble could be on its way as a result of the soaring dollar. This is something that certainly requires close watching now, as one (of many) potential triggers for pricking the bubble. The main fount of bubble conditions, money supply growth, still seems to be in fine fettle for the moment, but we know that the current boom is to some extent fundamentally different from a “normal” boom, in that it is far more concentrated in financial markets, so one will have to be careful with the interpretation of various data. We will discuss this particular aspect in greater detail in our next installment of the “Echo Boom” series (coming soon). Stay tuned.
The dollar index continues to go parabolic … to the chagrin of dollar borrowers not resident in the US – click to enlarge.