Latin America’s performance in the last few years has been dismal. With widening output gaps, central banks have started to loosen monetary policy. In contrast, the Central Bank of Mexico has been tightening monetary policy, despite low growth and deteriorating prospects. The bank faces a dilemma between using monetary policy counter-cyclically—as they did during the global crisis—or tightening it because of fear of a falling currency. They
have chosen the latter option.
On average, monetary policy implemented through a flexible inflation target allows both output and inflation to stabilize around their long-term values. This permits countercyclical monetary policy to be implemented, effective for softening a boom and containing losses in a downturn. This was a key element of the response of main Latin American central banks to the global financial crisis, which was weathered quite successfully (De Gregorio 2014). Monetary policy loosening played an important role, and that is one of the salient differences with their response and poor performance during the Asian Crisis (Alvarez and De Gregorio, 2014).
Between December 2008 and early February 2009, the central banks of Brazil, Chile, Colombia, Mexico and Peru started cutting rates to unprecedented levels. This was done in the context of currency weakening and high but declining inflation rates. Indeed, in November 2008, all these countries had inflation above 6 percent, well over their central bank targets for inflation. The forward-looking nature of monetary policy allows for anticipation of widening output gaps and a declining inflation outlook, with the consequent space for monetary stimulus. All of this is fully consistent with achieving the inflation target in the medium term.
Latin America’s performance in the last few years has been dismal. After a challenging episode of depreciation, low growth and inflation, the central banks of Brazil, Chile and Colombia have begun cutting rates, recognizing the need to add stimulus. In contrast, just the day after the Fed made the first hike in December 2015, the Mexican Central Bank (Banxico) raised the policy rate for the first time in several years, from 3 to 3.25 percent, and has steadily continued the tightening. Surprisingly, at that point, inflation–headline and core–as well as inflation expectations, were at their lowest ever. The pressures on the exchange rate front, although still on the weakening side, were diminishing (figure 1).
Output was growing moderately, as growth in 2015 was 2.6 percent, and in 2016, after tightening had started, about 2.1 percent. The first hike came when Banxico rescheduled its monetary policy meetings for immediately after those of the Fed. Clearly, they were playing Fed-follower. The monetary policy interest rate continued to rise up to 5.25 percent in October 2016, and in the aftermath of the Trump election, with additional tensions on the foreign exchange market, rates have risen to 6.25 percent as of early February.
Why is an inflation targeting central bank tightening monetary policy when a severe negative demand shock has hit the economy? Why run procyclical monetary policy? Is the Phillips curve gone? Consumer and business confidence dropped to record lows. There is a serious risk of facing trade restrictions from the United States and unstable foreign relationships with its most important neighbor and trade partner. Prospects for 2017 looks grim.
The Banxico survey reports median expected growth for this year at 1.6 percent, 0.6 percentage points less than before the US elections. The negative shock has been very serious. All news has been contractionary. Monetary policy in Mexico has been facing a serious dilemma: The need to support economic activity and the fear that if monetary policy is not tightened, and kept in line with the fed funds of the United States, Mexico could be subject to major dislocation in financial markets: rapid capital outflows and disorderly depreciation. Indeed, around the
first hike one of the main official arguments was about capital flows. This is what Vegh and Vulletin (2012) call Fear of Free Falling (FFF), which impedes many economies from avoiding policy procyclicality.
Why right after the Brexit vote and the fall in the British pound, did the Bank of England cut rates? Why did Australia, an advanced commodity exporter, cut rates from 4.75 to 1.5 percent starting in late 2011, with inflation above its target and against a backdrop of a 40 percent currency depreciation? Or why did Canada, another commodity exporter, cut rates in 2015 facing a depreciation? In Latin America, Chile also cut rates during 2013-14, with a significant depreciation and inflation above the target. These countries have been able to run countercyclical monetary policy because they do not suffer from the FFF and their target is medium-term inflation.
There is no evidence that not tightening will induce a disorderly adjustment in financial markets. It cannot be ruled out, but market fears are many times induced by policymakers’ fears. Moreover, across the region, including Mexico, sovereign spreads, CDS and EMBI+, had a mild increase after the US election results were known, which has
been mostly reversed. This is a symptom that depreciations have been more fundamentally driven than caused by overshooting or another financial anomaly. In the case of Mexico, a depreciation is needed just to sustain the existing savings-investment balance without activity bearing the burden of the adjustment.
Some may argue that tightening monetary policy is just accommodation to a large depreciation, since monetary conditions are not as contractionary as they seem just from looking at the interest rate. An index of monetary conditions would show an expansionary trend due to the weakening of the peso. These indices weight changes in interest rate with changes in the exchange rate. The relative weight of the exchange rate is the ratio between the sensitivity of aggregate demand to the exchange rate and the sensitivity with respect to the interest rate. For example, using weights as those of the Bank of Canada 3:1, a 1 percentage point of higher interest rate would be offset by a 3 percent depreciation. The monetary policy rate has increase by 3.25 percentage points, while the
peso has fallen, in multilateral terms, according to BIS estimates, by 17 percent.
However, to gauge the degree of expansion by looking at the monetary conditions index could be very misleading, in particular when the exchange rate has changed because its fundamental value has changed (see, e.g., Stevens, 1998). The monetary stance should be measured around fundamental values and not disregarding what has caused the exchange
rate change. Indeed, the aggregate demand should be estimated around equilibrium interest and exchange rates. These equilibrium values may fluctuate, in particular the exchange rate, which depends on external conditions. And the Mexican peso has moved as expected given the bad economic news, and consistent with a weakening of its equilibrium value.
If there are concerns about the consequences of a weak currency on financial stability or an unwarranted depreciation, central banks could start intervening in the foreign exchange market, as Banxico recently started doing, before hiking rates. However, the recent sharp depreciations in the region have shown that financial systems are resilient to exchange rate fluctuations.
In some circumstances, exchange rate depreciation may have inflationary consequences that may require a monetary policy response. This is particularly relevant in places with low monetary credibility, where the feedback from deprecation to inflationary expectations may be significant. These “second round effects” caused by the deanchoring
of inflation expectations may cause inflation to persist for a long period of time.
In those cases, preemptive early actions may avoid the need for a stronger policy stance in the future, with greater negative consequences on activity and employment. However, this reasoning cannot be generalized, especially in central banks that have a medium-term inflation target. Indeed, as Gopinath (2016) has reported for a large sample of countries that includes Mexico, the passthrough from exchange rates to inflation is front loaded, and hence, there are no relevant inflationary consequences at two-years horizon. This is also confirmed in De Gregorio (2016) for episodes of large depreciations, such as those experienced by most Latin American countries in recent years.
The last hike took place with the largest monthly increase in inflation in a very long time, (1.7 percent), and it could be justified, despite it was largely a one-off adjustment due to the liberalization in gasoline prices. Monthly core inflation was 0.5 percent. However, the entire process of tightening calls into question the argument that it has been driven mostly to anchor inflation expectations. Actually, it has been driven mostly by the fear of
free falling and accommodating its neighbor’s policy actions. Moreover, in late 2015, when tightening started, very few expected the political developments of its neighbor.
One-year ahead inflation expectations in Mexico have indeed been rising. The Banxico survey shows the median inflation for the end of this year at 5.2 percent against 3.3 percent at the end of 2015. In contrast, expected inflation for 2018 has only risen by 0.3 percentage points.
It is not only the level of the interest rate, relatively high, that matters for monetary policy, but also its changes. Signaling commitment with inflation when rates are relatively high may be costly for activity and medium-term inflation. Banxico is facing a serious challenge. Monetary policy has long and variable lags. To leave the current straightjacket it should focus communication and actions on inflation forecast in the medium term, say two years, and focus also on medium-term expectations. In most circumstances, monetary policy should be countercyclical, to contribute to macroeconomic stability.
Much progress has been made with independent central banks with a mandate on price stability. Now, it may be necessary to reduce excessive interference of exchange rate developments in the conduct of monetary policy, to get rid of the fear of free falling, and to strengthen credibility.
José De Gregorio
Universidad de Chile
Peterson Institute for International Economics
References
Alvarez, Roberto and José De Gregorio (2014), “Understanding Differences in Growth
Performance in Latin America and Developing Countries between the Asian and the Global
Financial Crises,” IMF Economic Review, Vol. 62, No. 4, pp. 494-525.
De Gregorio, José (2014), How Latin American Weathered the Global Financial Crisis,
Peterson Institute of International Economics, 2014.
De Gregorio, José (2016), “Large Depreciations: Recent Experience in Historical
Perspective,” PIIE Working Paper 16-8.
Gopinath, Gita (2016), “The International Price System,” Jackson Hole Symposium
Proceedings, Federal Reserve Bank of Kansas City.
Stevens, Glenn, 1998, «Pitfalls in the Use of Monetary Conditions Indexes,» Reserve Bank
of Australia Bulletin (August), pp. 34–43.
Vegh, Carlos A. and Guillermo Vuletin (2012), “Overcoming the Fear of Free Falling:
Monetary Policy Graduation in Emerging Markets,” NBER Working Paper No. 18175.
