Keynes's theory of liquidity preference and the behavior of US interest rates in the late 1920s and early 1930s
How does Keynes's theory of the bond market help contribute to an explanation of booms and busts such as the one that took place in the USA in the late 1920s and early 1930s? Since both Keynes's theory of the business cycle in chapter 22 and a sketch of a Keynes-Minsky cycle will be presented later, we limit our discussion here to a few points.
It is easy to tell a simple abstract story about how Keynes's theory of the interest rate operates in a boom. Rising bond prices in a general environment of investor optimism lead to the expectation held with increasing confidence by an increasing proportion of investors that bond prices will continue to rise or at least not fall significantly over the foreseeable future. Investors therefore are likely to shift portions of their financial wealth from money to bonds over time. This causes interest rates to keep falling and bond prices to keep rising. Falling interest rates help stimulate rapid growth in AD, creating or accelerating a boom in the real sector by making it more attractive for firms and households to engage in debt-financed spending. This will increase the financial fragility of firms, households, and banks. At some point, unexpected problems in the real and/or financial sectors create pessimistic expectations and possibly a collapse in confidence, kicking the expansion into reverse.
It is more difficult to tell a simple yet adequate story about the actual behavior of interest rate determinants in the bubble of the late 1920s by focusing exclusively on the bond market through the lens of chapters 1315. This is due in part to a complex relation between interest rates and stock prices in the boom that is not discussed in these chapters. Though it is assumed in the chapters that investors have to choose between money and bonds, in fact they can buy either stocks or bonds or both.
Long-term bond prices did rise by almost 20 percent in the boom of 1926 through early 1929. But stock prices rose by about 125 percent between 1926 and late 1929, generating increasingly optimistic expectations of capital gains on stocks held with increasing confidence. This induced many investors to sell bonds and buy stocks, which kept long-term interest rates higher (and bond prices lower) in 1927-1929 than they otherwise would have been. Meanwhile, there was an enormous increase in speculators' demand for short-term broker loans to buy stocks near the peak of the bubble.15 This caused a jump in short-term interest rates on broker loans - from 4 percent in early 1928 to 14 percent in mid-1929. The spread between long and short interest rates fell precipitously from 1928 through 1929, inverting the yield curve.16 Speculation-driven short rates actually moved above long rates in 1928 and most of 1929 (Federal Reserve Bank of Cleveland 1998, p. 199).17Now consider what might happen once a strong bond market bubble bursts, bond prices start to fall, and interest rates begin to rise. Expectations would shift rather quickly from optimism to pessimism about the future path of bond prices, triggering an increase in the rate of bond sales that would accelerate the pace of interest rate increases. This would generate even more pessimistic expectations. Because expectations of rising bond prices just prior to the start of the crash turned out to be profoundly mistaken, confidence in the expectation-formation process would be likely to collapse. This alone would cause an additional sharp upward pressure on the interest rate. Rapidly rising interest rates would accelerate the decline in AD caused by falling capital investment and deteriorating consumption spending, which would reinforce the ongoing rise in liquidity preference. Should a process of this kind take place at a time when firms, households, and financial institutions have developed fragile balance sheets, the financial system itself might face collapse, as it did in the USA in the early 1930s.