Only Unfunded Government Spending Can Save Us: Secular Stagnation and Currency Stimulus
June 30, 2015
Nearly eight full years have passed since the beginning of the Great Recession in the United States, and yet the national – and global – economic recovery continues to grind along lethargically. Growth has picked up domestically, especially relative to America’s international counterparts, but wage growth continues to be languorous and the US still faces significant structural challenges in trade and production. But the reason for the tepid recovery is no mystery. Economists such as Larry Summers, Paul Krugman, and the International Monetary Fund have pinpointed the problem: secular stagnation. What is secular stagnation, and why is increased government spending the only way to solve it? Read on for Wonk Tank’s explanation.
(Image: Economist Larry Summers, foremost modern advocate of secular stagnation theory. Credit: Wired Academic, Creative Commons License)
Secular stagnation, first conceptualized in 1939 as an explanation of the lasting effects of the Great Depression, is an outgrowth of Keynesian economic ideas that denounced the classical economic theory that markets will inherently self-correct (For a more detailed explanation, check out our fiscal policy primer).
In the same vein of thinking, but applied to a different set of economic phenomena, economist Alvin Hansen proposed the idea of secular stagnation. Hansen also argued that the economy could not always return to its optimal levels of output without government intervention – but blamed the phenomenon on naturally slowing rates of population and innovation growth, following the incredible economic expansion brought on by the Industrial Revolution and unsustainable population growth rates engendered by, among other things, mass immigration to the US.
Hansen’s theory was that companies and other supply-side economic actors see population and innovation rates naturally leveling off, so they reduce their investment levels relative to the social optimum because they anticipate lower returns in the future: a variation on the same concept as Keynes’ vicious cycle hypothesis. The policy prescription is also the same: governments must use the tools at their disposal to stimulate demand and bump up production levels and, correspondingly, employment and wages.
The following years neither fully vindicated nor impugned Hansen’s thesis. Despite the effectiveness of Keynesian-inspired New Deal efforts to increase employment and spur the economy, economists generally agree that it was really World War II that finally pulled the global economy out of the Great Depression. And World War II was exactly the kind of phenomenon Hansen argued would stimulate demand: the government engaged in massive and unprecedented spending and stimulus, production levels burgeoned enormously, and after the war, soldiers coming home fueled the Baby Boom generation and resurged population growth.
(Image: Federal Government Spending as a Percentage of GDP. Notice the massive surge during World War II, 1942-1945. Source: Federal Reserve Economic Data)
So while Hansen’s predictions failed to materialize, his underlying hypothesis has, empirically, remained relatively unassailed.
History Repeats Itself?
A growing number of economists, led by former World Bank Chief Economist and US Treasury Secretary Larry Summers, believe secular stagnation is making a comeback. The circumstances are remarkably similar to last time. The global economy is stuck in a (theoretically) cyclical rut brought on by a major economic recession; population growth is slowing; and technological progress is curtailing after a period of breakneck innovation brought on by things like computers, cell phones, and the internet. (This combines the idea of secular stagnation with the related technological slowdown hypothesis.)
Secular stagnation is so dangerous because it proposes that cyclical, impermanent stagnation could become structural, long-term and immutable economic slowdown – if private actors are not incentivized in some way to over-invest relative to their expected levels of return. The solution Summers proposes is the same as last time: juice the private sector by ramping up overall production levels through public investment, specifically substantial new government fiscal spending and monetary stimulation.
Nobel Prize-winning economist Paul Krugman has jumped in to bolster Summers’ case, and recently a report by the International Monetary Fund concludes that private investment is falling not because investors worry about low interest rates and foreign governments hoarding excessive savings (counterarguments advanced by former Fed Chair Ben Bernanke, among others), but rather because they anticipate problematically low demand – thus the secular stagnation hypothesis. The IMF’s recommendation is wholly unsurprising to those paying attention to the debate thus far: to jump start the economy again, governments must do more to stimulate aggregate demand.
The first and most obvious way to combat secular stagnation would be to induce a surge in population growth – accomplished most easily by increasing immigration flows into the US. But doing so would, of course, raise questions about resource availability and, more broadly, the capacity of the US to regularly handle population increases larger than the natural rate.
A better idea would be for the government to simply spend more money, increasing employment and aggregate demand.
There are obvious areas in which the government should be investing more into the economy. America’s level of public goods provision is, to put it diplomatically, severely lacking. The U.S. desperately needs more public investment in education, infrastructure, green energy initiatives, medical research and development, and other services – all of which require significant supplementary public expenditure.
Aside from the immediate benefits of this additional spending – better schools, better roads, cleaner energy and more effective medicine – the economic externalities are significant. Increased public investment would create countless new, well-paying jobs, dropping the unemployment rate. It would enhance domestic production levels, spurring economic growth. And these two factors would arguably lead to higher wages across the board as low unemployment gives workers more bargaining power to demand raises.
All of this is good, but it raises an important question. How could we possibly raise the revenue required to pay for any of this? And this is the most important part of the entire proposal: for at least part of this new spending, we shouldn’t.
Of course, proposals such as President Obama’s to repatriate and tax corporate profits held overseas, and then use that money to invest in infrastructure, are sound policy ideas that would do the country well to consider. But beyond that limited case in which firms have already accrued the profits under scrutiny, and are simply hiding from their collective statutory obligation, over-taxing creates disincentives to invest and economic deadweight loss. That is not to say distributional and public good priorities do not outweigh those disadvantages – just that heavy taxation is a suboptimal funding mechanism if other policy options exist.
Luckily, there is a better alternative. Instead of ensuring this massive new public expenditure is fully funded by revenue increases, the government should bankroll these new initiatives by simply printing money.
Printing Money Is Not As Dangerous As It Seems
The argument against funding government spending by having the Federal Reserve merely write the Treasury Department a big check is intuitive: rampant inflation would ensue. Injecting more money into the economy would just raise prices all around and, on net, solve nothing.
Inflation is not a bad thing, however. In fact, it is actually necessary for economic growth. Money is productive: more money means more employment, more spending, and more demand. The Fed wants the money supply to increase annually; it targets a two percent annual inflation rate when setting monetary policy.
That said, runaway inflation is definitely bad. Prices spiraling upwards and out of control, exceeding real economic growth, would be disastrous. And at first glance, a new money printing initiative of this scale would directly lead to that eventuality.
But the reality is not so simple. In fact it is, and has been for the last several years, entirely feasible for the Fed to print massive quantities of new money without inflation exceeding the target two percent rate. The following chart shows exactly why:
(Image: The United States Monetary Base, M1 Money Stock – in other words, the money supply, and the Personal Consumption Expenditure Price Index – a measure of inflation, indexed so that their levels in September 2012 are equal to 100. Source: Federal Reserve Economic Data)
To summarize the chart, there are a few concepts to understand. The monetary base is what the Fed controls and can expand or contract with monetary policy; it represents the base level of money in the economy. The money supply, here represented by the M1 money stock, is the total value of what we can think of as “usable money” in the economy: the total currency in circulation, of course, as well as the value of liquid assets that can be converted easily and quickly to cash – such as bank deposits. (This differs from the monetary base because of the money multiplier, which is best explained here.)
The third line in the chart above is the Personal Consumption Expenditure Price Index, a measure of inflation that tracks the price level of consumption expenditures (I use it here as opposed to the far more popular Consumer Price Index because CPI tends to overstate actual inflation, and the Fed tends to prefer PCE as a metric for policymaking).
In theory, the monetary base, money supply, and price level should increase at roughly the same rate. And we can see that prior to the recession, the three measures tracked each other extremely closely. But after the Fed lowered interest rates to zero at the end of 2008 in a successful effort to expand the money supply, the monetary base and money supply increased at roughly similar rates and far outpaced inflation – likely because poor economic conditions held prices down as people simply could not afford higher priced items. Thus, the recession proved that in times of economic downturn, expansionary monetary policy could absolutely be successful in helping the economy recover without skyrocketing prices.
In late 2012, however, we see a significant divergence. In September of that year, the Fed was looking to further expand the money supply to accelerate the recovery, but had already lowered interest rates to zero. (Notably, economists have recently begun to push back against the “zero bound” idea that central banks cannot further expand the money supply through conventional interest rate policy after already pushing rates down to zero, but that is an article for another time.)
So the Fed embarked on a round of Quantitative Easing (QE): a more ambitious, controversial, and direct expansion of the monetary base where the central bank began to purchase bonds and other securities of corporations and financial institutions – the idea being that this influx of cash a firm receives upon selling a security would be used by these entities to make investments in labor, capital, etc., that would increase employment and stimulate the economy.
The Fed kept its QE program in place for about two years, before ending the latest round in late 2014. Whether QE “worked” or not in the broader sense of being good for the economy is the subject of significant debate, but it is hard to argue with the results: expansionary monetary policy kept the economy moving in a positive direction and recovering from the supply side (albeit slowly), and kept the unemployment rate moving lower, even as relative fiscal austerity (outside of the stimulus bill which created more than three million jobs) hampered growth from the demand side.
Just as importantly, it in no way led to the runaway inflation critics feared, as the chart above shows: the price level did not rise any faster than it had previously – in fact, it rose more slowly. Lastly, QE devalued the dollar’s exchange rate internationally, which made U.S.-produced goods cheaper and boosted export levels.
If QE was so successful, then, why can it not be the solution to secular stagnation? Why should we abandon it and focus on even more risky and unconventional methods of monetary expansion – in this case proposing to write a check directly to the federal government?
Quantitative Easing Is Inefficient
The answer, again, lies in the chart above. Beginning when the Fed embarked on its new QE program, there is a significant divarication in the chart between the monetary base and the actual money supply – the base increases at a significantly faster rate. But again, all else being equal, the monetary base and M1 should increase at the same rate.
That they bifurcate instead is because QE does not involve the Fed injecting money directly into the economy. It involves the central bank directing currency into the financial system – into corporations, banks, and other financial intermediaries. In other words, it represents classic supply side, trickle-down economics: it assumes that firms will use their new capital to make productive investments and hire workers.
But remember the premise of secular stagnation: that firms will not, on their own, make the socially optimal level of hiring and investment because they expect that economic stagnation will prevent them from getting the returns they need to justify investment costs.
So when the Fed used QE to, for example, create new money for commercial banks to lend out, many banks simply ended up keeping the money as excess reserves:
(Image: The Monetary Base, Money Supply, and Excess Reserves Held by Financial Institutions, all indexed such that their values are equal to 100 at the beginning of the Fed’s latest round of QE. Source: Federal Reserve Economic Data)
This also helps to explain why inflation remained so low: the injection of money into the financial system did not directly translate into a shot in the arm of the actual consumer economy writ large. As a result, aggregate market supply and demand did not change as much as would be predicted by a simplified model, and accordingly prices did not move much.
In fact we can calculate just how inefficient QE was. If M1 had increased at the same rate the monetary base did as a result of the Fed’s expansionary policy, it would be approximately $697 billion larger than it is. That is nearly $700 billion of pure economic deadweight loss, or almost the entire cost of the American Investment and Recovery Act (also known as the stimulus).
Therein lies the problem with QE and trickle-down economics more generally: while the government can put all the money it wants into financial instruments, there is no guarantee that the money will actually be used in productive ways.
But the Fed can ensure that money it injects into the system is used productively if, instead of just providing additional liquidity to the financial system, it simply writes a check to a government that will use that money for public goods provision. And that is the whole concept behind fixing secular stagnation.
Relying on private investment to fix a structural disincentive for private investment is circular reasoning. Public investment – the kind that provides public goods and services that can be used by anyone, good-paying jobs, and positive externalities such as better medical care, education, and energy systems – is the solution, and that is where the Fed should direct its stimulus efforts.
How Much Funding Can the Fed Provide Without Triggering Mass Inflation?
To quantify this, we have to start off with a few assumptions that, while likely unrealistic, will help to bound the results. First, assume that the money multiplier is equal to its average over the ten years before the beginning of the recession and expansionary monetary policy, which is approximately 1.77. That is to say, assume that an additional dollar added by the Fed to the monetary base would increase the actual supply of liquid money by $1.77. This is unlikely, as the money multiplier actually fell to less than one after the crisis as financial institutions prioritized stability over growth – but since we are proposing bypassing the financial sector, it is worth ignoring those changes.
Second, we will assume that inflation increases on a one for one basis with the money supply: that a one percent increase in M1 would lead consumer-facing prices to also increase by one percent. This is a highly unrealistic assumption; the data we have examined thus far indicates that inflation remains muted despite monetary distension, and prices are sticky, which is to say there are costs to changing price levels that help slow down inflation.
Nevertheless, it is a useful assumption when attempting to create a model that shows the worst-case scenario – meaning the scenario in which the Fed can print the least money without wreaking economic havoc.
(Image: Annual inflation as measured by the Personal Consumption Expenditure Index from the beginning of the Fed’s QE program through May 2015, as well as the less volatile Core PCE Index, which does not factor in food and energy prices that are more susceptible to fluctuation. Source: Federal Reserve Economic Data)
Remember that the Fed targets a 2% inflation rate. Since QE began, inflation has consistently been below that rate, meaning there has been a ton of room for the central bank to add money to the money supply. Revisiting our previous assumptions – inflation increases one for one with the money supply – and using the average of PCE and core PCE, we can map the potential versus actual increase in the monetary base and money supply using the following equation, where the inflation target is 2% and the money multiplier is 1.77:
Subtracting the actual monetary base from the potential monetary base will benchmark approximately how much more money the Fed could directly inject into the economy without inflation rising above its target level for each year since expansionary monetary policy began with the drop of interest rates (Note: For the purposes of calculating potential monetary base only, projections are cumulative, i.e. each year’s “actual” monetary base in the equation above is calculated as the previous year’s potential base plus the actual percentage change in the base):
The results confirm our expectations: since the start of the financial crisis, the Fed could have printed nearly $200 billion in additional money and simply written a check to the federal government to spur additional public investment. And this is likely a massive underestimate: given how unresponsive inflation has been to expansion in the money supply, and if the Fed used this money printing program in place of QE, rather than in addition to it as this model predicts, the actual amount of monetary stimulus could be orders of magnitude larger.
And there is no reason the Fed cannot start now. Inflation shows little sign of speeding up, especially with a Fed rate hike expected at the end of the year. The economy is still recovering anemically. Private investment is still not at optimal levels. The dollar continues to be over-valued abroad, dragging on exports. Just as importantly, American infrastructure is decaying, our education system continues to lag behind the rest of the world, and healthcare is still inefficient and over-priced.
Secular stagnation is very scary concept for economists: it is structural, not cyclical, it cannot be traced to any particular policy decision, and it will not inherently self-correct. Fighting it requires innovative and ambitious policy tools – tools which the American government unquestionably possess.
We cannot keep blaming the lugubriously recovering economy on the continued cyclical effects of the Great Recession; rather, we have to recognize the structural and long-term implications of a suboptimal level of private investment because of insufficient private incentives.
Like it or not, the government has a role that it is uniquely suited to play.
Luckily, the government also has the economic flexibility to play it. Inflation remains repressed no matter how expansionary American monetary policy has been, and attempting to inflate the money supply through the financial system has proven inefficient.
Desperate times call for unorthodox measures, and the threat of a structural economic slowdown unquestionably qualifies as one of those times. The Fed needs to think outside of the standard monetary policy toolbox, and use their monopoly over currency as the power that it is.
The message to Janet Yellen and her Board is simple. Write a check to the Treasury Department. Invest in public goods provision. Create jobs, raise wages, supercharge domestic production. Or give in to inflation hawks, and let the middle class and broader economy suffer as a result.
The choice is clear.
Attached here is a document with all of the data and calculations used in this article.
For more on the dispute over secular stagnation between Larry Summers and Ben Bernanke, check out this The Economist aggregation.
Further reading on some policy proposals put forth to increase America’s public goods spending:
- Senator Bernie Sanders (I-VT) has proposed a $1 trillion new spending spree on infrastructure.
- A report by the American Energy Innovation Council calls for a tripling of the $5 billion the government spends annually on energy research.
- Forbes’ Steven Salsburg advocates doubling our biomedical and basic science research expenditures at a cost of nearly $40 billion.
Additional Blog Posts
Student Blog Disclaimer
The views expressed on the Student Blog are the author’s opinions and don’t necessarily represent the Penn Wharton Public Policy Initiative’s strategies, recommendations, or opinions.