Are We Headed for Financial Repression?

Thinking about the path of U.S. debt and the likely outcomes

You are reading Economic Forces, a free weekly newsletter on economics, especially price theory, without the politics. Economic Forces arrives weekly in the inboxes of over 20,000 subscribers. You can support our newsletter by sharing this free post or becoming a paid subscriber:

---

Much of the political discussion over the last week or two has revolved around the “One Big Beautiful Bill.” I don’t particularly care to get into the particulars of the bill. Instead, I want to focus on a growing realization that seems to have been triggered by the bill and what this means going forward.

If you look at government debt-to-GDP ratios for most countries for which there is considerable time series data, you see similar patterns. There are large, quite dramatic spikes around significant events, like wars or the Great Depression, followed by relatively slow, gradual reductions over time. The following chart on U.S. government debt as a percentage of GDP from the Congressional Budget Office illustrates this well.

Readers will notice a change in this pattern that occurs around the time of the financial crisis of 2008-09. In the period since, the U.S. government debt-to-GDP ratio has increased quite dramatically for a prolonged period of time. One might attribute this is a series of negative shocks: financial crisis, a slow recovery, and a global pandemic. I’m not inclined to believe that view. I think that there are broader, structural forces that are driving this trend. However, let’s entertain this “series of unfortunate events” take on the rise in the debt. If this is true, then one would expect to see this dramatic increase in the debt-to-GDP ratio peak and gradually decline. However, this is not what anyone is projecting, including the Congressional Budget Office.

In short, U.S. government debt is on an unsustainable path. The U.S. now spends more money on interest payments on the debt than it does on national defense — and many consider the national defense budget bloated!

With debt on an unsustainable path, the U.S. government effectively has three options. The first is to impose fiscal discipline. This would require some combination of higher taxes, a reduction in spending, and entitlement reform.

A second option would be for the government could simply print money to buy back the debt. In this U.S., the government would not — and arguably could not — pursue this strategy deliberately. The Federal Reserve is largely independent of the political process and is tasked with price stability. However, the Federal Reserve is also tasked with maintaining moderate long-term interest rates. In the event that investors lose faith in the U.S.’s ability to repay its debt (in real terms), they are likely to start fleeing U.S. Treasury securities. Volatility in the Treasury market would likely lead to intervention by the Federal Reserve to “calm the Treasury markets” with the net result being that the Federal Reserve’s balance sheet expands to hold a greater proportion of U.S. government debt.

Finally, the federal government could resort to financial repression. By financial repression, what I mean is either an implicit or explicit requirement of the U.S. government that financial firms hold a greater amount of government debt than they would without such policies in place.

The significance of the commentary around the One Big Beautiful Bill seems to be that there is a growing recognition that the U.S. government lacks a commitment to imposing fiscal discipline.

Given the trajectory of the debt and the lack of fiscal discipline, it seems that inflation and/or financial repression are likely consequences. Are we headed for financial repression?

The Bank of England

While contemporary discussions of central banks often focus on the technocratic nature of monetary policy, frequent readers of Economic Forces know that I like to discuss the fiscal origins of central banks. The origins of the Bank of England are particularly relevant.

Following the Glorious Revolution and the ascendancy of William III to the throne, England was thrust immediately into war with France. Louis XIV of France supported the return of James II to the throne in England. William III, on the other hand, saw an opportunity to use his new role to forge a European alliance as a counterbalance to France.

This conflict created a number of issues, in particular with regards to financing. England lacked the fiscal capacity to raise sufficient revenue and, according to Lord Macaulay the taxes being levied at this time were already historically very high. Borrowing was difficult. The English Civil War, the repudiation of the debt following the Stuart restoration, and the threat of restoration of the James II didn’t make that any easier.

As a result, in 1694, William III’s whig supporters in Parliament chartered the Bank of England in exchange for the bank lending 1.2 million pounds to the crown at an 8 percent interest rate. Given England’s fiscal limitations and recent turmoil, these seem like extremely favorable terms.

Why does this matter?

One reason that this matters is that this example fits with our definition of financial repression. Given the favorable terms, it is almost certainly the case that this arrangement required the financial system to hold more public debt than it would have otherwise.

The second reason that this matters is that it can help us think about the underlying economic incentives associated with this arrangement. While most people have focused on the role of the Bank of England in financing the public debt, Earl Thompson emphasized a characteristic of this scheme that seems particularly important. Notably, Thompson argued that this arrangement created something tantamount to a modern poison pill strategy.

In corporate finance, poison pills are used to protect companies from hostile takeovers. The restoration of James II would be the equivalent of a hostile takeover in this context. The first thing that any successful revolution does is repudiate the debt of the previous regime. A sovereign-debt holding bank at the center of the financial system therefore acts as a poison pill since the restoration of James II and a repudiation of the debt would come at a significant economic cost because it would create financial turmoil.

At the same time, this scheme addresses the commitment problem. Given that the crown lacked the ability to make a credible commitment to repay the debt, this form of financial repression essentially creates a type of credibility. It aligns the shareholders of the bank with the interests of the crown. It also creates a steep consequence should the king default because doing so would significantly disrupt the financial system.

Finally, this commitment isn’t costless. If the bank was chartered without the government funding requirement, it likely would have allocated its resources toward private, productive uses. Thus, in a sense, the king (or perhaps just as accurately, Parliament) was effectively purchasing commitment via a reduction in private investment relative to the counterfactual.

These lessons from the Bank of England carry over more generally.

When is it optimal for states to resort to financial repression?

Given this discussion of the Bank of England, one might wonder how this relates to more general cases. Fortunately, V.V. Chari, Allesandro Dovis, and Patrick Kehoe have an excellent paper that formalizes the incentives for what they call “optimal financial repression.”

The basic premise of their model is as follows. Suppose that there are two types of debt holders: households and banks. As long as the government can credibly commit to repaying its debt, there is no problem. On the other hand, when the government cannot credibly commit to repay its debt, its incentives to default are going to depend on the distribution of debt between groups.

Suppose that all of the debt is held by households. If the government decides to default on the debt, this is really just equivalent to a one-time lump sum tax on households holding the debt.

On the other hand, suppose that at least some amount of government debt is held by banks. Now, a default on the debt reduces the net worth of the bank. In doing so, this will have the effect of reducing investment. A reduction in investment will have a much larger effect on economic activity by making future economic output permanently lower than it would have been otherwise.

Of course, it is possible to engineer a lump sum tax on households even if banks are holding the debt. The way to do that is through bank bailouts. The government could default on its debt and bail out the banks. In doing so, this prevents the net worth of banks from declining and thus prevents the resulting decline in investment. However, this also means that the gains from default are much lower since this is equivalent to a targeted default. Households pay a lump sum tax, but since they only hold a fraction of the debt, the benefits of the default to the government are much smaller.

In this context, it is fairly straightforward to understand why governments might resort to financial repression. If the government cannot credibly commit to repaying its debts, then it could effectively “purchase” some degree of commitment by requiring that banks hold a larger amount of government debt on their balance sheets than they otherwise would. The cost of purchasing this commitment is that there will be less private investment in the economy than there would be otherwise. Nonetheless, since everyone knows that the benefits to default decline with the fraction of debt held by banks, requiring that banks hold larger amounts of government debt reduces the benefits from — and thus the likelihood of — default.

What would financial repression look like?

Given that the U.S. federal government seems to lack a credible commitment to repaying its debt, it seems much more reasonable to predict that the U.S. debt problem results in financial repression than tax increases or entitlement reform.

Financial repression in the United States would be nothing new. The National Banking Acts, for example, required banks to purchase U.S. Treasury bonds to finance the Civil War. The cost was that banks effectively had to “back” their note issuance with these bonds, which created a relatively inelastic currency. Even current bank regulation provides implicit subsidies for banks to hold U.S. Treasury securities on their balance sheet.

A move toward explicit financial repression could come in the form of simply imposing a formal requirement that banks hold a certain percentage of their assets in U.S. Treasury securities.

Another possibility, as mentioned by people like Charles Calomiris, would be to have the Federal Reserve eliminate the interest paid on reserves. Currently, the Federal Reserve holds a large amount of U.S. Treasury securities on its balance sheet and pays interest on reserves. When the interest earned on Treasury securities exceeds the interest paid on reserves, the Federal Reserve earns a profit, most of which is returned to the Treasury Department. By paying interest on reserves, this reduces the interest spread between the Fed’s assets and liabilities (and following the recent bout of inflation, actually turned negative).

Eliminating interest on reserves without any corresponding change would likely be inflationary given that it would lead to a significant reduction in the demand for reserves and thus a dramatic expansion of broader measures of the money supply. As a result, reserve requirements would need to increase since, by doing so, banks would not be able to reduce their demand for reserves to the degree that they would like.

Currently, both interest on reserves and reserve requirements are tools of the Federal Reserve. Absent a significant bout of inflation, the Federal Reserve is unlikely to institute these on their own. However, Congress could always changes the statute to eliminate the Federal Reserve’s authority to pay interest on reserves and declare that the Federal Reserve can determine reserve requirements above some explicit, statutory amount.

This would be an indirect form of financial repression. Higher reserve requirements when reserves pay no interest is akin to a tax on banks. The “revenue” from this tax would come in the form of the Federal Reserve maintaining its ability to have a large balance sheet. The Federal Reserve would then be capable of holding large amounts of U.S. Treasury securities that earn interest while borrowing from banks at zero percent interest. This would allow the Federal Reserve to position itself as a buyer of last resort for U.S. Treasury securities while attempting to minimize the inflationary impact of doing so.

Are these really all of the options?

A much rosier scenario would be one in which the development of artificial intelligence creates a substantial increase in productivity. In terms of the debt-to-GDP ratio, nominal GDP growth is unlikely to change as a result of AI since, all else equal, increases in real growth will tend to be accompanied by lower rates of inflation. However, rapid productivity would generate more tax revenue from the same tax base. In doing so, all else equal, the rising tax revenues would reduce budget deficits. As a result, debt would grow more slowly over time and perhaps find a more sustainable path.

Of course, “all else equal” is doing a lot of work in these descriptions. It is also possible that rising tax revenues convince policymakers that they can spend even more money — and they might be tempted to do so as a result of any societal disruptions brought about by AI and the transition to new types and forms of production. It is also possible that a central bank focused on inflation will maintain its target rate of inflation even in the face of rising productivity. This, however, would make the debt-to-GDP ratio decline faster than increased revenues alone. However, how much faster is likely to depend on the costs associated with the erosion of information conveyed by relative prices when the central bank tries to maintain an inflation target in the midst of rapidly rising productivity.

Some Final Thoughts

Unless something rather significant changes, U.S. government debt will continue on this unsustainable path as a simple matter of arithmetic. It is therefore important to consider what that means going forward.

Predictions of politicians “waking up” to the problem seem quite naive. The international monetary system allows the U.S. government to borrow more cheaply than it otherwise would. The role of the federal government has greatly expanded to the point that the only way to reduce spending is through significant entitlement reform. There is no political appetite for that.

Similarly, there is no political appetite for the across-the-board increases in taxes. Politicians are still promising to pay for things by “taxing the rich,” which would provide insignificant revenue in comparison with the magnitude of the debt.

Predicting a likely outcome leaves us with one of two possibilities — or a combination thereof. Those likely outcomes are a significant period of high inflation and/or financial repression. Given the relative independence of the Federal Reserve, it is unlikely that we will observe a coordinated effort to pay down the debt through inflation. The Federal Reserve, however, might be forced to intervene in Treasury markets, the result of which will be higher inflation.

But a likely outcome is financial repression. This is a tool that states have been using for centuries. It is naive to think that the U.S. will not attempt to use this tool in the event that politicians perceive a looming debt crisis. And, unlike the other options, the connection between the actions of the government and the costs are largely hidden from the public.

Or maybe AI will save us.