By Andrés Fernández, Daniel Hernaiz and Andrew Powell
Most of the largest economies in Latin America have adopted inflation targeting . A huge advantage of having such an anchor, and not relying on a fixed exchange rate to curb diverging expectations, is that the exchange rate is then determined by the market and can adjust to shocks. Given the large shocks the region has been subjected to of late, this has proved very useful and exchange rates have been very, very flexible—much more so than many might have thought possible given the recent financial history. The large currency depreciations between 2014-2015 cushioned external and internal shocks, but they also generated some inflationary pressures; pass-through was not zero although it was also less than might have been anticipated.
The reaction of the central banks was then to defend the inflation target and maintain the credibility of the nominal anchor that had allowed such flexibility. This has now created the conditions for monetary authorities to lower policy rates and contribute more, together with exchange rate movements, to the stabilization of economic activity. But we argue below such monetary accommodation should be gradual, and that the gains from a less restrictive stance may be limited, especially if the region faces further adverse external shocks.
The recent 2017 Latin American and Caribbean Macroeconomic Report analyzes the challenges faced by central banks. The report highlights three principle messages.
First, inflation seems to be under control in the majority of countries, independently of the exchange rate regime (fixed, floating, intermediate), although there are some exceptions, such as the case of Venezuela. Figure 1 highlights inflation dynamics in six central banks that pursue inflation targeting and shows that the trend of rising inflation between 2014-2015 was reversed, and by the beginning of 2017 average inflation is again within (or close to) the target range. Moreover, the figure indicates that inflation expectations have been converging once again towards target levels.
Fig 1: Inflation Rates, Targets and Expectations in Latin American Countries with Inflation Targeting Regimes
Source: Latin American Macro Watch and Central Banks’ Websites
Note: In the case of Peru, the graph illustrates inflation expectations each September for the subsequent year.
This should allow central banks to lower policy interest rates, and that is exactly what most have been doing as shown in Figure 2. Four of the five central banks from the largest economies have lowered or, at least, left their policy rate constant over the last three months. The exception has been Mexico. But Mexico suffered a new external shock as a result of the election result in the United States and the possible renogotiation of NAFTA. The depreciation of the Mexican peso has been very significant indeed, again acting as a shock absorber for the economy, and so the Bank of Mexico reacted to manage the potential impacts on inflation and on inflation expectations.
Fig 2. Monetary Policy Rates in Latin American Countries with Inflation Targeting Regimes
But how much monetary space is there? Not that much is the report’s second broad message. This can be seen in Figure 3, which depicts output (horizontal axis) and inflation (vertical axis) gaps for these countries between 2012-2016 (the solid lines). On one hand, output gaps have deteriorated, falling from a positive average gap of around 3% in 2013 to a negative one of 1%. This means that in just 4 years the average volume of production in these countries went from significantly above to significantly below the historic trend. Or, in other words, production is now below what it would be under normal conditions with the resources these economies have. This would seem to call for a more expansive monetary policy to soften the contractionary phase of economic activity.
Fig. 3 Average Output and Inflation Gaps in Latin American Countries with Inflation Targeting Regimes
Source: Authors’ calculations.
Note: The alternative scenarios are described in the text. The output gap for 2017 was computed as the average cyclical component of a Hodrick-Prescott decomposition (lambda=1600) applied to the seasonally-adjusted quarterly GDP forecast of the model. The inflation gap for 2017 was computed as the difference between the model’s forecast for inflation and the current inflation target
However, the Figure also shows that despite the progress of 2015-2016 to close the inflation gap —defined as the difference between observed inflation and the target—the gap is still positive at around 1.5 percentage points. Central banks face a dilemma with positive inflation gaps, although closing, and negative output gaps, that seem to be on the rise. In this “complicated quadrant,” the monetary policy tradeoffs are tricky: lower the rates faster to help stabilize economic activity at the possible cost of reversing progress on inflation, or take a more gradualist approach. What should central banks do?
The third message is that the gradualist approach is likely best. But why? The recommendations stem from analyzing simulations from a Neo-Keynesian monetary model calibrated for five economies. In particular, five forecast scenarios are considered. They are characterized in Figure 3 (the dotted lines). The first, called non-conditional, supposes that the average gaps close according to the inertia in the data, otherwise known as an unconditional forecast of the model. In this scenario, the inflation gap closes rapidly; the output gap less so, though it moves in the right direction (grey dotted lines). The second scenario supposes further adverse shocks, mostly of external origin: low commodity prices and higher international interest rates. In this case, both gaps worsen: the inflation gap widens again, and the output gap continues its march into negative territory (blue dotted line). The last three scenarios suppose that the fiscal and monetary authorities change the policy rules that they have historically followed as a response to the adverse shocks considered in Scenario 2. In Scenarios 3 and 4, the central banks decide to adopt a less restrictive position relative to the historic rule they have been following, or, in other words, a deeper cut in interest rates given a more negative output gap. Finally, Scenario 5 supposes that fiscal adjustment is delayed leaving less space for cuts in interest rates.
The results do not favor a change in the estimated historic rules. Scenario 5 is the least favorable of all, given that the two gaps worsen with respect to Scenario 2; the inflation gap increases; and the output gap turns even more negative. Scenarios 3 and 4 achieve only a moderate output stabilization, while the inflation increase is particularly large. Thus, the emphasis should be on the gradualness in the lowering of the policy rates in this new policy stance, especially if the region confronts adverse external shocks. Why is the less restrictive policy so ineffective in closing output gaps? The intuition is that the private sector realizes that the central bank will be less restrictive and so will increase demand—relative to the central bank maintaining the historic rule. This means inflation will be higher than otherwise and the central bank is forced to hike rates anyway. Another way to say this is that pass-through would rise. So, the result is higher inflation but little gain in terms of output. Gradualism then seems to be the best approach.
For economists, the last few years in Latin America have been an absolutely fascinating experiment in how inflation targeting regimes work in small and open economies subject to large shocks. Just as material scientists study metals under extreme stress so that we can then travel safely in modern airplanes at 500 miles per hour at 16,000 feet, it’s always useful to monetary policy where the shocks are larger and the tradeoffs are more acute. In Latin America, floating exchange rates have been very flexible and have acted as very useful shock absorbers. Central banks have been forced to raise policy rates procyclically because currency depreciations have resulted in some (moderate) pass through, but labeling this as procyclical monetary policy misses the point. The monetary stances (taking into account the movement of the exchange rate and not just the policy rate) have surely been more countercyclical. Our analysis of monetary policy in the region has just scratched the surface. There is much more to learn. We hope this blog will stimulate interest in the monetary policy dilemmas of these economies and may result in a new emerging economy theory of inflation targeting which seems to be lacking in the literature to date.
 In the 2016 Latin American and Caribbean Macroeconomic Report a medium pass-through of around 16% is estimated.
 There is still not that much impact on exports valued in dollars, but there was much more general impact on import penetration in the region – see http://www.iadb.org/en/research-and-data/2017-latin-american-and-caribbean-macroeconomic-report,20812.html