Stress Tests Can Help Determine How Much Capital Central Banks Need

This article is published in association with IMF.

Romain Veyrune

Central banks pride themselves on being data-driven and consensus-based. Surprisingly, though, there is little consensus among policymakers on a crucial issue: how much capital should these institutions hold?

Unlike commercial banks, central banks don’t have universally prescribed minimum capital requirements. They can’t go bankrupt, because they can always issue their own currency to meet their nominal payment obligations. Even so, a weak capital position can diminish institutional credibility and potentially increase risks to independence. Hence central banks do care about maintaining sufficient capital buffers. But there is a diversity of views on how to get there.

This is why we propose a new approach of stress-testing central banks to help them maintain a sound financial position.

Balance sheet risk received little attention through most of the long history of central banks. Before the global financial crisis, they typically had small balance sheets and were almost always profitable. This reflected that currency, their main liability, paid zero interest, and that they could invest the proceeds that they received from issuing currency in interest-bearing government bonds. Much of the profit was paid out as dividends to governments.

But the issue, which may sound arcane, has much more practical relevance today when central banks have taken on much more balance sheet risk, including using large-scale asset purchases to spur a faster recovery from the GFC and the pandemic.

Managing risk

This additional risk has translated into sizable losses, as they bought long-term bonds at low yields and eventually had to raise interest rates sharply. While the losses are not a good measure of the social value of central banks actions, which shortened the recessions and improved financial stability, they underscore the need to consider carefully how to better manage balance sheet risk.

Studying the bylaws of central banks provides little clarity about how to proceed. Many set their authorized capital as a fixed amount, which loses relevance over time due to inflation. Only a few institutions adjust their capital—based on inflation or gross domestic product—to keep it relevant.

Unfortunately, existing laws about distributing central bank profits are also quite mechanical. In some cases, they determine exactly how much gains should be kept or shared, which can result in having too much or too little capital. At best, these laws require banks to keep profits until they reach a minimum level of capital. But legal targets vary a lot—from 8 percent to 20 percent of base money—and there is little explanation for thresholds. At the other end of the spectrum, some central banks lack specific rules on capital, leaving it up to their boards to decide how to handle risks. Yet, whatever they choose to do, central banks are generally reluctant to explain their approach to a broader audience.

There’s a better way. The key is to ensure that capital cushions are more consistent with “policy solvency”—with the central bank’s ability to fulfil its mandate in an environment of much greater balance sheet risk. This means considering several factors, such as institutional objectives and activities.

Shock absorber

Specifically, stress-testing can help a central bank gauge the level of capital that would allow it to absorb large but plausible shocks without pushing capital to very low levels that could weaken its credibility and independence. To this end, IMF staff developed a quantitative model, building on 2015 research by Robert E. Hall and Ricardo Reis, that allows assessing how capital would evolve in a framework that takes account of interest rate risk, credit risk, and foreign exchange risk. A stress test would consider inflation and other broad economic dynamics, and how they would affect capital.

This approach can also help decide when a capital increase through profit retention is warranted—or when and how to share profits while protecting capital levels. Some central banks may find such a risk-based approach appealing, especially if they perceive that a weak capital position could constrain their independence. Others may view little risk to their credibility or independence, and prefer to retain their current policy for capital distributions. But even in these circumstances, they may see stress-testing as a way to enhance transparency about the likely effects of balance sheet actions such as quantitative easing, and desirable for strengthening public accountability.

The IMF has published a guidance note on central bank stress-testing, and we provide technical assistance on that topic to our member countries. To be sure, these unique institutions have a specific public mission which distinguishes them from commercial firms. But sometimes the approaches taken to supervise private banks can illuminate the debate about central banks.