Would a Gold Standard System Have Worsened the 2008 Financial Crisis?
August 30, 2012
(This item originally appeared at Forbes.com on August 30, 2012.)
A friend tells me that the soft-money advocates will always argue something along the lines of: “What if the financial crisis of 2008 had happened and we were on a gold standard?” They imply that a gold standard system would have made the crisis much worse.
Well, what if?
All of these kinds of soft-money arguments tend to boil down to the notion of devaluing the currency during a time of crisis. The actual strategy could be an overt currency devaluation, or it might be related to very low interest rates, or printing an excessive amount of money to deal with one problem or another, leading to a decline in currency value. A gold standard system would prevent a currency from declining in value; and thus, any excessive money-printing of this sort.
Let’s take, for purposes of example, a period from the end of June 2008 to the end of June 2009. The U.S. stock market had not yet begun to fall rapidly in June of ’08; indeed, the S&P 500 finished August at about the same level as it ended June. By the end of June 2009, the S&P500 was already recovering rapidly from its low in March 2009.
The value of the dollar was $930.25 per ounce of gold at the end of June 2008. At the end of June 2009, it was $981.75 per ounce. Thus, on a point-to-point basis, the dollar was not devalued during this crisis period. Indeed, if the U.S. had been on a gold standard system at the time, let’s say at a parity of $1000/oz., the outcome would have been about the same on a point-to-point basis. The value of the dollar would have been $1000/oz. at the end of June 2008, and the same $1000/oz. at the end of June 2009.
The dollar did not remain at an even value throughout the crisis period. Nor was it devalued, as the Keynesians like to imply. No, the dollar went up! In October of 2008, the dollar hit a high of $712.50/oz. (London PM fix basis). It then fell back. This spike in dollar value was reflected also in exchange rates with other currencies. The Federal Reserve’s broad trade-weighted dollar index went from 95.76 at the end of June 2008 to a high of 112.49 in November 2008, before falling back to 105.35 at the end of June 2009.
The basic reason for the financial crisis of 2008 was bank insolvency. Many banks had made too many loans to borrowers who could not pay them back. The prospect of widespread and chaotic bank default loomed. This spike in dollar value, in the middle of the crisis, just added a new problem to an already problematic scenario.
A gold standard system would have prevented this spike in dollar value. The value of the dollar would have remained at a stable $1000/oz. throughout the crisis period. If other countries were also using gold standard systems, then the exchange rates between their currencies and the dollar would have remained fixed and predictable, instead of varying wildly.
Why did the dollar spike higher in value during this time? The simplest reason is that demand was greater than supply. Beyond this, we have to hypothesize a little bit.
Dollar base money consists of two elements: banknotes and coins, and bank reserves recorded as deposits with the Federal Reserve. Because bank reserves do not pay interest, banks have tended, over time, to reduce their holdings of bank reserves. This increases profitability. However, having less reserves also increases banks’ dependence on short-term borrowing via the money market to deal with liquidity needs. As the crisis intensified, and indeed in the years since that time, banks found that they could no longer borrow from other banks easily or reliably. Banks didn’t trust each other anymore, because they knew that many banks had insolvency problems. Thus, banks wanted to keep more cash on hand, in the form of Federal Reserve bank reserve deposits.
Bank reserves thus expanded, from about 5% of base money outstanding to about 50%. This is a big move, but it actually just brought banks back to where they were in the 1950s, when bank reserves accounted for about 40% of base money. As a percent of deposits, bank reserves increased to about 15%, again roughly where it was in the 1950s.
Banks “demanded” this increase in a forthright way: they borrowed it voluntarily from the Fed, via the discount window, at interest. Thus, it is fairly safe to say that this represented a legitimate increase in base money demand. No wonder the dollar’s value was rising during that time, until this demand was met by the Fed.
A gold standard system is intended to increase supply to meet increased demand, so that the currency maintains a stable value. If the dollar has a tendency to rise, as it did in late 2008, a gold standard system would have automatically increased the monetary base to keep the value of the dollar at its gold parity.
If this increase in the monetary base, via the normal operating mechanisms of a gold standard system, was for some reason not sufficient, the Fed could have also acted as a “lender of last resort” in the 19th-century meaning of the term, or what I have called “19th Century Central Banking.” In fact, this is what the Fed did, via its discount window lending. This action is entirely consistent with a gold standard system, and mirrors what the Bank of England did in the 19th century while keeping the British pound pegged to gold.
Discount window lending is intended to be short-term, to respond to short-term changes in the need for bank reserves, which historically were closely related to the harvest season. However, the dramatic increase in banks’ desire to hold reserves has been a persistent pattern for some years now. With a gold standard system, these short-term loans would eventually be replaced with a longer-term adjustment in base money outstanding, likely through purchases of either gold bullion or Treasury bonds. In fact, this is exactly what Ben Bernanke did with “QE1,” which replaced discount window lending with MBS on the Fed’s balance sheet.
Many other things were done during this crisis period. Banks such as Washington Mutual, Countrywide Financial, Merrill Lynch and others were restructured or merged. AIG was bailed out by the federal government, thus indirectly bailing out AIG’s creditors. TARP was passed. Fannie Mae and Freddie Mac were effectively nationalized. Citigroup was recapitalized by the federal government. A blanket guarantee on money market funds was applied, and many other additional steps were taken.
Whether these were good ideas or bad is a topic for further discussion. However, one thing they have in common is: they could have all been done within the context of a gold standard system. Indeed, the most effective action, corresponding to the stock market bottom in March 2009 and its quick recovery afterwards, was a ruling by the FASB that banks were allowed to mark their assets “to model,” which is to say, make up their values. The Fed had nothing to do with this.
Thus, we find that what the Fed did during the crisis of 2008 was, in fact, much like what would have happened with a gold standard system. If anything, a gold standard system would have operated more quickly and more smoothly, supplying additional base money as soon as the dollar began to rise in value, an indicator of increased base money demand. Also, a gold standard system would have maintained the dollar at an even value, for example a parity of $1,000/oz., throughout the period, thus avoiding the additional turmoil of currency instability piled upon the existing problem of bank insolvency.
The Fed was created for exactly this purpose. It was always intended to operate alongside a gold standard system, and did so for 58 years, from 1913 to 1971. In the Federal Reserve Act of 1913, it states: “Nothing in this act … shall be considered to repeal the parity provisions contained in an act approved March 14, 1900,” which referred to the Gold Standard Act of 1900. It was only in later years that the Keynesians managed to turn the Fed into a supposed all-purpose economic Mr. Fixit via the magic of monetary distortion.
Actually, with a gold standard system, it is likely that the Fed could have been much less involved in 2008. The reason banks had a sudden need for more reserves in 2008 is that they began the crisis with so little. During the gold standard era, from 1865 to 1971, banks consistently held much larger reserves, on the order of 30%-50% of total base money outstanding. The reason that banks were holding such small reserves at the beginning of the crisis is related to their expectation for constant, daily Fed involvement in money markets, which has become a characteristic of the past few decades of the floating fiat dollar. If banks had started the crisis with their historically much higher reserve levels, then there might not have been a need for Fed action at all.