«Extraordinary times call for extraordinary measures». That's how simply Ben Bernanke summarized it in February 2009 after declaring that the Federal Reserve would do everything possible, within its limits, to restore financial stability and economic prosperity. Four years later, Mario Draghi followed suit with his now famous «believe me», endorsing an undertaking the ECB had gradually assumed since the start of the crisis. Briefly put, those in charge of conducting the monetary policy of the main advanced economies (also of the Bank of England and of Japan) have proved their will to use all their ammunition, including the extraordinary measures mentioned by Bernanke, to support their respective realms. But, no matter how laudable the intention (and it is), and no matter how effective the measures (more questionable),(1) this approach is not without its risks.

To begin with, the unconventional measures resorted to recently by the large central banks are still that: not very conventional.(2) As such, in many respects they are still in a trial phase, somewhat similar to a drug that works on paper but has yet to be tested thoroughly enough to be administered with the utmost confidence. In such a case, we all know that it would only be used in situations of extreme urgency and if there were no other alternative. Sound familiar? In fact, what encouraged this «new» monetary policy was the onslaught of an extreme crisis when the traditional defenses had been practically disarmed.

Continuing with our simile, the main risk lies in a lack of experience regarding the right dose and duration of the remedy: interrupting treatment too soon could lead to a relapse; but excessive ingestion (in terms of size or duration) could cause addiction, adverse reactions or unwelcome side effects. Given the customary bias in managing economic policy and a series of aggravating factors that range from fear of relapse to government financing needs, stimuli are more likely to be withdrawn too late rather than too early.

Let us start, then, with the effects of addiction: should the current huge amounts of liquidity persist, this might generate what is known as a moral hazard, insofar as it feeds the expectations (of financial institutions, the public Treasury or whoever benefits from the current monetary facilities) of being able to rely on this kind of support in the future. At the same time, it might also lead to dependence on central bank financing and prevent the interbank market from getting back to normal, supplanting the usual market structures for managing liquidity with injections by the lender of last resort.

Even if addiction isn't created, an excess of stimuli could be toxic for the patient (in this case, the economy). On the one hand, the direct and massive purchase of assets by the main central banks could possibly distort credit assignment, artificially pushing up the price of certain assets and helping to form speculative bubbles, both domestic and foreign. On the other hand, such an extraordinary expansion of easy monetary conditions such as the present could end up de-anchoring inflation expectations, pressurizing prices and ultimately wiping out the fundamental aim of monetary policy: price stability.

If this were a common drug, our simile wouldn't work anymore; so let us return to the unconventional: let us suppose that this is a radioactive treatment, whose handling requires careful precautionary measures so as not to contaminate the person administering it. Because, in fact, pushing monetary policy to the limit involves risks for the central bank itself. By expanding or changing the composition of its balance sheet, the monetary authority is altering its financial risk profile, exposing itself to potential deterioration, both in its asset position and in its reputation.

It essentially incurs three kinds of risk: interest rate, credit and, ultimately, a loss of credibility. Acquiring bonds whose price falls when the interest rate goes up (for example, sovereign bonds) could incur losses for the central bank due to changes in value (for the mark-to-market part of its portfolio) or could result in these assets being sold at a loss. On the other hand, it is also subject to credit risk resulting both from the direct purchase of assets (for example, mortgage-backed securities in the case of the Fed or sovereign bonds acquired under the ECB's Securities Markets Program) and also from providing credit to banks in exchange for collateral (for example, the ECB's LTROs). In both cases, the monetary authority could incur losses if there is any default on payment, at the level of asset or collateral.

In any case, these central banks' exposure to credit risk is relatively low as significant measures have been taken to protect themselves against such eventualities, such as imposing haircuts on collateral to absorb potential losses in value or requiring, in general, good quality assets both in direct purchases (the case of the mortgage-backed securities acquired by the ECB) and also in collateral. Moreover, it is important to remember that a central bank can operate with negative capital. In other words, if an extraordinarily adverse scenario did occur that entailed such huge losses for the central bank they would ultimately eat up all its assets, even then it could continue to operate, in the short term, without too many problems.

Although it's true that the operational significance of this negative capital would be irrelevant to some extent, it is also true that, if it ended up restricting the central bank's independence or marred its credibility (conditioning the management of monetary policy), it might be inconvenient. In other words, in principle, negative capital could be restored via successive profits or by recapitalization on the part of the public Treasury. However, the possibility of a central bank having to resort to such Treasury recapitalization might be perceived as a loss of independence. On the other hand, resorting too much to seigniorage(3) could undermine the credibility of the central bank's commitment to price stability, de-anchoring expectations and generating inflation. Nevertheless, I repeat: it is a risk but not a very likely one. This loss of credibility would be more likely to come from a late withdrawal of stimuli or the excessive relative weight of certain assets within the central bank's balance sheet, compromising price or financial stability.

In short, minimizing the aforementioned risks requires the careful calibration and planning of the withdrawal of stimuli, combining them with other economic policy tools. In this respect, good communication of this exit strategy would help, conveying confidence and safeguarding the central banks' invaluable credibility. At the end of the day, the fact that a drug is under trial does not mean it doesn't work but, given the risks involved, it is always much better to be safe than sorry.