The following is excerpted from the November issue of the "Fundamentals" newsletter published by Research Affiliates. To read the full commentary, click here.
For the 50 years from 1951 through 2000, U.S. GDP growth averaged 3.3% per year. We can attribute this historical growth to three primary components: 1.4% from population growth, 0.3% from a rising employment rate, and 1.6% from growth of output per person employed (productivity). In the coming 20 years, all three components of growth will be much lower.
Births and fertility rates are declining. Immigration has slowed to a trickle in response to harsh immigration policies and a dimmed growth outlook. For the next two decades, the U.S. population will grow by only 0.7% per year, half the rate of growth witnessed in the late 20th century.
The total employment rate will continue to decline as boomers move from their 50s into their 60s and 70s. Whereas a rising employment rate added 0.3% per year to GDP growth from 1950 through 2000, the demographic effect on the employment ratio will subtract 0.2% for the next two decades. The population will be growing by 0.7% per year, but the employable work force will be growing by only about 0.5% per year.
Productivity growth faces severe headwinds from both demography and fiscal contraction. If the productivity growth of our aging labor force is 0.5%, then future GDP growth will be centered around 1%.
Demography guarantees a sharp slowdown in GDP growth. 1% real growth is still growth. It's a joy to behold, if our expectations are anchored on zero, as was the case throughout human history before the industrial revolution. But, if our expectations are tied to the 3% growth from 1950 to 2000, the 1% growth seems abysmal.
Slower growth is not a serious problem; it's the expectations gap that poses economic and political dangers in the years ahead.
Demography is Destiny?
Demography is destiny. But, while it is the most immutable force shaping our future, it is not the only force that matters.
This 1% “new normal” for GDP growth is, of course, subject to considerable uncertainty. Our policy choices will have a dramatic influence on all three drivers of economic growth. Immigration reform, while politically difficult, could help slow the decline in population growth. We can increase the rate of employment by revising our transfer payment policies to provide ample incentives for employers to employ and for the labor force to seek employment. We can boost productivity by changing our tax policy to encourage savings and investment, rather than printing money to support debt-financed consumption.
It seems unlikely that we can reverse the decline in fertility rates observed across the developed world.4 Immigration can, however, move the needle on U.S. population growth by several tenths of a percentage point per year. The United States absorbed one million immigrants per year in the past when our total population was much smaller than today. The positive contributions of immigration to growth are well documented, as summarized by Professor Gordon Hanson (2012) at UC San Diego. While the economics are uncontroversial, the politics are not promising. Both the right and the left have their own separate reasons to resist changing our immigration policies.
The structure of our tax and transfer payment systems has a significant influence on employment rates. Our combination of regressive payroll taxes and phase-out of transfer payments results in effective marginal tax rates above 40% — in some cases far above 40%—for the working poor (Kotlikoff and Rapson, 2006), while we tax carried interest in private equity deals at only 15%. As the authors observe: “The patterns by age and income of marginal net tax rates on earnings, marginal net tax rates on saving, and tax-arbitrage opportunities can be summarized with one word—bizarre.” Reform of our tax and transfer payment systems, to reward work instead of penalizing it, could add as many as 30 million jobs in a few short years (Arnott, 2011b).
The most important determinant of our productivity per person is the amount of capital we have available for investment, and wise use of that capital. Investing to improve productivity in all of its forms, from machine tools, to transportation infrastructure, to education, can all raise our productivity. Investing requires sav¬ings and we have been saving far too little. To increase productivity requires that we reorient our economy away from debt-financed consumption and toward saving-financed investment.
In this context, ending the mortgage interest deduction seems oddly absent from our political debate, perhaps because it would run counter to long-standing bipartisan policies promoting the American dream of home ownership and perhaps because it would simply be too unpopular. Nonetheless, the tax deductibility of interest on home mortgages (which was effectively reaffirmed when the Tax Reform Act of 1986 rescinded the deductibility of credit card interest) has been credibly identified as one of the factors that eventuated in the financial crisis (Szegö, 2011). Because borrowing against one's house is inexpensive on an after-tax basis, it was easy to rationalize converting the wealth effect of rising prices from a psychological phenomenon to actual cash on hand for consumer spending. Banks and families are still paying for the good times, and the overhang of properties in foreclosure ensures that the residential real estate market will not fully recover for years to come. Yet there is little discussion of changing the tax code with regard to debt collateralized by borrowers' homes.
Unless we change our policies to encourage immigration, employment, and investment, our new normal growth rate will be 1%. Government tax and spending plans, based upon the Rosy Scenario of extrapolating past growth rates, are likely to exacerbate our already unsustainable deficits and dangerous accumulation of debt.
Christopher J. Brightman is head of investment management at Research Affiliates.