Is the Fiscal Theory of the Price Level a solution? Here are my open questions.
Back in 2019 and 2020, many people were talking about Modern Monetary Theory (MMT) as the new paradigm for fiscal and monetary policy that explained why governments could run large deficits for a very long time without causing inflation. I have written several posts about this theory explaining why I think it is a flawed or even ahistorical theory. Well, since then things on the inflation front have changed and MMT has moved back into the shadows. But now a new theory emerges as an interesting alternative to the quantity theory of money or Keynesian theories of inflation: The Fiscal Theory of the Price Level (FTPL).
Before I discuss FTPL, I would like to make a disclaimer. I am a total novice in understanding this theory and its implications. Most of my knowledge is based on an excellent introduction written by Thomas Coleman, Bryan Oliver and Larry Siegel. It is written for generalist investors, and I highly recommend it. But because I am a novice, my questions and criticisms below may be totally flawed and wrongheaded. So, I encourage my readers to get in touch with me to tell me where I am wrong, what I don’t understand or why FTPL might not be subject to the problems I raise below.
With this disclaimer in place, let me give you the shortest possible explanation of FTPL I can think of. At the core of the theory is an extension of the good old quantity theory of money.
Basically, FTPL says that in our modern world it is not just the quantity of money in circulation that matters for inflation, but also the quantity of bonds. FTPL correctly recognizes that both money and government bonds are liabilities on the balance sheet of the government that need to be covered with present and future income from taxes and other sources of revenue. The quantity of money and/or bonds in the economy can rise without creating inflation if investors believe that the government will cover these additional liabilities with future government surpluses.
To assess whether the government can credibly pay back all its liabilities in the future, FTPL calculates the fiscal capacity of the government as a discounted present value of future cash flows. In essence, the government is treated like a company where instead of equity value and dividend payments, one tries to explain the present value of government liabilities and the interest paid on them.
The beauty of FTPL is that it can neatly explain why there was no inflation after 2008 even though the amount of government bonds issued by the US government skyrocketed and even though money in circulation increased as well. FTPL simply states that fiscal austerity by the government after the end of the financial crisis created the expectation that the government will be able to offset new liabilities by increasing future tax revenues and reduced deficits.
And though it is early days, FTPL indicates that the current increase in inflation is unlikely to persist. The US government as well as other Western governments are reducing their fiscal deficits and are projected to do so in 2023 and 2024. Additionally, the growth in M4 seems to be peaking indicating that the monetary base will not continue to grow at the fast pace we have seen in the last two years. These and other signs would lead a believer in FTPL to conclude that we are dealing with a one-off inflation spike that should normalise over time.
But whatever the current implications of FTPL are, while I like some of the concepts like the explicit connection between monetary and fiscal policy, I think FTPL makes three key assumptions that I would challenge:
Investors are weighing current government actions against the long-term expected behaviour of the government.
Investors believe the government when it promises to offset current spending with future budget surpluses or higher tax revenues.
Investors are rational.
Richard Thaler likes to joke that only two out of the following three statements can be true at the same time:
Investors are rational
Markets are efficient
Financial intermediation is 9% of GDP
Similarly, I think only two out of the three key assumptions of FTPL can be true at the same time.
As for the first point, it is easy to point out that if you ask random people on the street or even random people on a trading floor how they think about inflation, very few if any would come up with an explanation that they look at a spreadsheet that contains a discounted cash flow model of the government and its future liabilities. If you are wondering what such a spreadsheet would look like, take a look at this paper which has done all the work needed to test FTPL.
Reading this paper shows that the Congressional Budget Office (CBO) currently projects no government surplus until 2051 (when current projections end). As the note shows, in order to cover the current and future liabilities of the US government, it needs to run a surplus of 3.1% of GDP from 2052 onwards into eternity. I don’t know about the early years of the United States but with the help of the macrohistory database I could go back to 1870 and look at US budget surpluses and deficits. It turns out that over the last 150 years, the US government only once managed to achieve a surplus of more than 3%. In 1948 the government surplus was 4.3% of GDP. Other than that, the best the government could come up with was a 2.3% surplus.
This brings me to the third assumption of FTPL. I don’t know about you but if investors really look that far into the future and trust the government to run surpluses of more than 3% into eternity starting in 2052, you have to be either irrational or fantastically credulous. But I certainly would doubt that investors who believe these two points are rational and rationally change their assumptions based on current developments.
Similarly, if I assume that investors are rational and take into account the very distant future and its implications for today, they shouldn’t trust the government to offset current liabilities with future revenues. As a result, they should demand extremely high risk premia on US Treasuries.
Or alternatively, one may assume investors are rational and they trust the government to offset current liabilities with future surpluses, then they are clearly not paying attention to the long-term forecasts of the CBO.
Don’t get me wrong, I find the concepts in FTPL intriguing. If you ask me, the way forward would be to give up the assumption of rationality. If we start to treat government cash flows like the cash flows of a company, then I would expect that the present value of these future cash flows is subject to the same kinds of irrational behaviours we witness in stock markets.
Inflation would thus become subject to potential bubbles and crashes as investors and businesses exhibit herding behaviour in the form of price gauging (everybody expects inflation to rise, so businesses increase their prices and consumers demand higher wages, thus creating an inflation bubble). We would also expect to see inflation crashes (e.g., after a hyperinflation when a new currency is introduced). And we would have to manage inflation not by manipulating interest rates but by manipulating investors’ confidence and trust in government just like a company tries to influence its share price by engaging with investors on its investment plans etc. And even then, we would have to admit that we have very little if any control over inflation, nor do we really know how inflation is driven. And coincidentally, that is something that seems to become clearer and clearer every day.