Jason Furman and Larry Summers begin a recent paper with the sentence "The last generation has witnessed an epochal decline in real interest rates in the United States and around the world despite large buildups of government debt." Their paper provides many useful insights, such as the effects of longer retirements and lower productivity growth on interest rates, but I think they're getting one core idea wrong. It seems increasingly probable that interest rates have declined because of large buildups of government debt. Here's why.
In 1971 President Nixon suspended dollar convertibility to gold, placing fiat currency at the top of the money pyramid, and more specifically the US dollar at the top of the global money hierarchy. Whether for reasons of habit or pragmatism -- people still had tax bills coming due in dollars -- the switch was accepted and the dollar became, for many intents and purposes, gold. Before 1971, when the US government issued more debt, interest rates tended to rise because investors knew that there were more claims on the government's gold reserves. This increased the odds of default or a devaluation of the dollar with respect to the money at the top of the pyramid, gold. Today when the US government issues more debt, it is literally expanding the money at the top of the pyramid -- in a sense, it is creating more gold.
The effect on interest rates from an increase in money at the top of the pyramid is different from the effect from an increase in money one or two rungs down the pyramid. To illustrate, say the government gives $1,000 to every American, issuing $330 billion of debt to do it. Econ 101 will tell you that this increases the demand for loanable funds, thus raising the price of money (interest rates). What it sometimes forgets to mention is that there is now an additional $330 billion in people's bank accounts, as well as an additional $330 billion of bank reserves. The effect on interest rates depends on what people do with that money. They can leave the money in the account, pay down debt, or transfer it to someone else in exchange for goods, services, or assets.
To take a simple case, if everyone leaves the extra $1,000 in their account, interest rates will decrease. The government can simply offer the $330 billion of Treasuries to the banks, which have an extra $330 billion of reserves. As long as the Treasuries pay slightly more than the interest that the Federal Reserve pays on reserves, banks will take this deal (ignoring bank liquidity and reserve regulations, which don't change the key point and in the case of reserve requirements are now zero). The government sets the interest rate on reserves and, by extension, Treasuries. Of course, if everyone leaves the money in their account, policymakers will concurrently decrease rates in an attempt to stimulate the economy.
Now, not everyone will leave the money in the account. If it is used to pay off debt, interest rates will rise (repaying debt destroys money / decreases bank reserves). If it is used to buy something, the deposits created in one account are transferred to another and the overall amount of bank reserves remains the same. The pressure on interest rates then depends on whether the money is spent on goods and services or a financial asset. If it is used to buy goods or services, like a car or a cab ride, there will be upward pressure on rates -- policymakers won't have to cut rates if people are consuming, since this means businesses are profitable and probably investing with an optimistic eye to the future. If the money is used to buy a financial asset, like a share in a mutual fund, there will be downward pressure on rates -- money flowing around the financial system is good for people who receive crumbs from the slices being passed around but generally does not spur consumer confidence and business investment. On net, if money goes to people who need it for paying off loans or buying things, there will be upward pressure on rates. When money goes to people who don't use it on goods and services, there will be upward pressure on asset prices -- or, to say the same thing, downward pressure on interest rates -- as money sloshes around the financial system. In this case, high asset prices are less a sign of business dynamism and more a result of financial system dynamics.
Nowadays, the issue with more government debt isn't higher interest rates, it's a question of what people are doing with their money. The decline in rates since the 1980s is not so much a sign of an aging population and slower technological growth as it is the result of vast amounts of money lying dormant in a few bank accounts. Low rates are not necessarily bad in themselves, but their symptoms, such as increased financial risk-taking, and causes, such as democracy-fraying levels of inequality, are real concerns. Trying to fix the symptoms without addressing the causes would be dangerous.
