The idea of secular stagnation dates to Alvin Hansen, who wrote in the 1930s that it was possible for private-sector investment demand to fall far short of private-sector saving and that interest rates could not fall far enough to bring them into balance. As a consequence, Hansen argued, there might be a permanent need for large government budget deficits if the economy is to avoid protracted weakness and deflation.
World War II and the postwar spending boom lifted the U.S. economy from the stagnation of the 1930s, leading some to conclude that the secular-stagnation hypothesis was wrong. An alternative view that I think is closer to right holds that the hypothesis was correct but that sufficient remedies were found initially in wartime military spending, then in massive national projects such as building out the suburbs and reducing saving by allowing Social Security to meet retirement needs and making consumer credit widely available.
Fast-forward to the present. In the industrial world, real interest rates average well below zero and markets suggest they will remain below zero for decades to come. The ratio of government debt to gross domestic product is very high and rising rapidly. And yet even with super-stimulative monetary and fiscal policy, growth is sluggish at best and, even a decade after the financial crisis, inflation has not, on average, risen to central banks’ 2 percent target.
The core of the problem is that there is not enough private investment to absorb, at normal interest rates, all the private saving. The result is extremely low interest rates, weak demand, and low growth and inflation, along with the bidding up of the price of existing capital assets.
What has happened to private saving and private investment? Many things, including increases in saving caused by people having fewer children, more inequality, longer retirement periods and increased uncertainty. Probably more important, demand for private investment has fallen off as the economy’s structure has changed. Computing power costs a tiny fraction of what it used to. Malls have been replaced by e-commerce. People prefer small urban apartments to large suburban houses. Cars and appliances need to be replaced less often. In any event, the end of labor force growth means less demand for new capital.
This all has been widely recognized in discussions of the falling neutral real rate of interest, something the Fed has taken aboard with Chair Jerome H. Powell saying that the current real interest rate of about 0.5 percent was very close to the neutral rate, a decline of perhaps 3 percentage points from neutral real rates a generation ago.
Our paper takes matters further by pointing out that real rates would have declined by far more than they have but for increases in government debt, pay-as-you-go pensions and public health insurance. Using a variety of modeling approaches, we estimate that but for public policy, neutral real rates would likely have declined by as much as 7 percent since 1970 and would be several percentage points negative today.
Some key aspects of our analysis include:
- If the neutral real rate is significantly negative, there can be no assurance that even in the long run economies will reach full employment without appropriate policy. The traditional Keynesian view in which permanent depression is possible is more right than the New Keynesian approach in which unemployment is attributed only to temporary price rigidities.
- The large increases in government debts that we have witnessed in recent decades are less a consequence of fiscal irresponsibility than a response to shortfalls in private investment relative to private saving. For the world as a whole, it is a matter of arithmetic that the government can run a deficit only to the extent that the private sector runs a surplus. Rising private-sector surpluses have been a driver of public-sector deficits.
- Satisfactory growth may, given the economy’s current structure, depend on unsustainable policy settings. Witness the slowdown likely in the United States when budget deficits stop rising, even though they are at what may well be unsustainable levels. Likewise the adequate growth of the 2005-2007 period required low rates, an epic housing bubble and vast erosion of credit standards.
- If full employment is to be maintained in the years ahead, it will be necessary to assure that investment absorbs saving. Low interest rates may be insufficient to assure that this occurs and may have various toxic effects, including bubbles in asset prices and misallocations of investment.
- Fiscal measures, including larger budget deficits, improved Social Security that reduces retirement saving, redistributions of income to poorer consumers with higher spending propensities, and social insurance that reduces precautionary saving, all operate to reduce saving and raise neutral real rates.
- Measures such as reductions in monopoly power, investment mandates (such as the retirement of coal-fired power plants) and investment incentives by operating to increase investment also operate to raise neutral real rates.
We think our results, if they stand up to professional scrutiny, will require a rethinking of conventional views regarding macroeconomics and macroeconomic policy.