You can set your watch by the Swiss trains. They are the envy of the world in many ways. The secret to the system is not just high-quality engineering and scrupulous maintenance. It is also the fact that the schedule is the same every day of the year, and that if one train is late (a rare occurrence that elicits a public apology), no other train waits for connecting passengers. Everyone believes that the trains will run on time because they do.
Credibility is at the core of central banking as well. When a credible central banker speaks, households, businesses, and governments listen. And, because they expect the banker to do what she says, their response – measured in terms of how much they work, save, and invest – will reinforce the outcome policymakers seek.
But credibility is tough to earn and – as the Swiss National Bank (SNB) recently learned when it ended a three-year commitment to prevent a rise in the franc versus the euro – easy to lose.
It is like the trains – if they frequently miss their schedule, you will have to make other plans. In the same way, to be credible, central banks must commit to a strategy that works over time. Most important, people must believe that the central bank will not renege on that strategy in the future. When, instead, future incentives to renege are expected to be strong, the commitment lacks credibility. Economists say that such plans lack time consistency.
Time consistency is a key issue for central banks in many arenas. One example is the role of monetary policy independence. Making a central bank independent enhances the credibility of its commitment to keep inflation low. Research shows that independent central banks deliver low inflation with a smaller sacrifice of output and employment.
Time consistency is the Achilles heel of fixed-exchange rate regimes. In a world of free capital flows, central banks can either fix their exchange rate or they can pursue a discretionary monetary policy, but – at least over time – they can’t do both (for more on this “impossible trinity,” see here). The promise to fix an exchange rate means that the public, through its demand for currency, determines the size of the central bank’s balance sheet, which can no longer be used to set a policy interest rate.
At the end of World War II, when capital controls were widespread, so were fixed exchange rates. By the start of this century, with capital controls largely eliminated (aside from key exceptions like China), most economies had shifted from a fixed to a floating exchange rate regime.
The reason is that doubts frequently arise about a central bank’s promise to refrain from exploiting monetary policy discretion. Such doubts invite speculative attacks, making the fixed-exchange rate promise very costly or impossible to maintain. As a result, in a world of free capital flows, fixed exchange rate regimes are famously fragile.
A classic example is that of the United Kingdom’s 1992 exit from the European Exchange Rate Mechanism (ERM). In effect, the ERM fixed the U.K. pound at an overvalued level to the German Deutsche Mark at a time when Britain was in deep recession and Germany was experiencing a post-unification boom. Knowing that the U.K. government would be reluctant to let interest rates rise (to defend the fixed exchange rate) for an extended period, speculators attacked. In fact, the Bank of England’s rate hike on September 16, 1992 – “Black Wednesday” – accelerated the attack, because speculators knew that a sustained tightening of monetary policy would deepen the recession.
Trying to defend an overvalued currency, as the U.K. authorities did, invites speculative attack. Speculators know that the central bank only has a limited volume of foreign currency to sell to maintain the peg. Raising interest rates to make the currency attractive can work temporarily when business cycles are well aligned and other conditions (such as a robust financial system) warrant, but eventually the fragility is revealed.
It is easier to sustain an undervalued exchange rate because the central bank only needs to buy foreign currency, not sell its inevitably finite holdings. That is all the more so when cyclical considerations – such as a weak economy – or falling prices warrant the monetary expansion associated with foreign currency purchases.
This brings us back to Switzerland. What new lesson do we learn from the SNB’s policy reversal? The answer is that even fixing an exchange rate at a lower level by purchasing foreign currency in a weak economy with falling prices can create sufficient doubt to make the strategy time inconsistent. Put differently, even a pledge to buy a foreign currency without limit may lack credibility.
Back in September 2011, the SNB sought to counter a “massive overvaluation” of the franc that posed “an acute threat to the Swiss economy” and “the risk of a deflationary development.” Accordingly, it set a floor of CHF1.20/€, promising to “buy foreign currency in unlimited quantities.” Translation: The SNB would no longer use discretionary monetary tools (like interest rates) to set policy, but would let its balance sheet be determined by the desire of the public for Swiss francs at the exchange rate floor of CHF1.20/€.
For more than three years, the pledge worked, with the exchange rate settling slightly above CHF1.20/€. But achieving this goal required the SNB to purchase an enormous quantity of euros. Foreign exchange reserves soared from CHF255 billion in August 2011 to CHF462 billion as of November 2014, the latest available data. This surge accounted for more than all of the rise of SNB assets over this period (see chart). The foreign currency holdings alone represent more than 70 percent of Swiss GDP (currently about CHF 650 billion).
Swiss National Bank Assets (Billions of Swiss Francs), 2011-November 2014
Source: Swiss National Bank.
From an economic perspective, the policy was consistent with cyclical conditions and with price stability. Deflation abated, moving from -0.7% in 2012 to -0.1% in 2014, while growth remained between 1 and 2 percent. Forward-looking prospects also remained consistent with policymakers maintaining the currency floor: for example, the IMF World Economic Outlook in October 2014 projected 2015 Swiss economic growth of 1.6% and inflation of 0.2%, both slightly higher than 2014.
So, what drove the SNB to abandon its commitment? What was the source of the time inconsistency? In our view, it was political forces wishing to limit further SNB balance sheet expansion. Popular concern about potential losses to taxpayers from the purchase of euros was sufficient to trigger a November 2014 referendum that would have forced the SNB to increase its gold holdings and effectively constrain currency holdings. While the Swiss overwhelmingly defeated the referendum, the controversy nevertheless affected the currency policy: speculators knew that a large further increase in the SNB balance sheet would be politically risky, casting doubt on the sustainability of the policy commitment.
Subsequent developments – like the Greek crisis and the prospect of ECB quantitative easing – persuaded speculators that further SNB euro purchases would be needed to maintain the floor. And, the amounts could be large. Had the SNB policy commitment been more credible, these flows might have been smaller, but the irony of speculative attacks is that they can become self-fulfilling. In this case, ever-larger positions in the euro would have increased the SNB’s eventual losses in the event of a policy reversal, raising doubts about the willingness to purchase euros. As it is, the 17 percent surge of the currency to CHF1.00/€ (at this writing) that followed the currency policy reversal may have cost Switzerland more than 10 percent of GDP in losses on SNB currency holdings.
Yet, the aggregate economic losses from a sustained bout of unexpected deflation could be even larger. Swiss prices already are falling (albeit partly due to the welcome plunge in oil prices). The currency surge will hit import prices directly and domestic prices through the shock to aggregate demand (Swiss trade in goods and services accounts for more than 130 percent of GDP).
The bottom line: even when it makes considerable economic sense, a fixed exchange rate can be difficult to sustain. We knew this was true when the objective is to prevent one’s currency from depreciating. The lesson of the SNB’s experience is that it can be very difficult to thwart appreciation as well. Even the Swiss, with their uncanny ability to make the trains run on time, couldn’t do this.
Keep that in mind if, for example, Greece were to leave the euro area. Once a country exits, the Maastricht Treaty notion of an irrevocable commitment to the single currency would be lost, converting the monetary union to just another fixed-exchange regime. If this triggers renewed worries about “currency redenomination risk,” that regime could prove fragile as well.