One of the common perceptions about emerging markets (EMs) is that inflation rises sharply when currencies weaken, because weaker currencies push up domestic prices of imported goods and services. This phenomenon has a name: FX pass-through.
Yet currency weakness can equally be deflationary. Outflows can slow down economic growth and thus weaken inflation by pushing down the prices of non-tradables, turning the conventional thesis of FX pass-through on its head.
We examined data on FX pass-through for the countries in the JPMorgan GBI EM Global Diversified index, a widely-used benchmark. Using both country-level and index-level data, we find that FX pass-through is a myth.
In fact, we find exactly the opposite of the FX pass-through thesis in that periods of weaker currencies are consistently associated with lower inflation, while periods of stronger currencies are consistently associated with higher inflation.
The evidence at index level is incontrovertible. We weighted EM countries’ currency and inflation indices by their respective index weights for the full life of the index (Dec 2003 through Dec 2016).
As the chart below shows, EM currencies appreciated by 46 per cent between 2003 and 2008 followed by a long period of meaningful depreciation (43 per cent) between 2011 and 2015. These two episodes were briefly interrupted around the time of the Developed Market Crisis, during which currencies fell and then recovered sharply.
What is particularly revealing about this chart is that each of the distinct periods of alternating currency strength and weakness have consistently been associated with higher and lower inflation, respectively. Inflation rises when currencies go up and inflation falls when currencies go down. The conventional thesis of FX pass-through predicts exactly the opposite behaviour.
Specifically, inflation increased steadily during the bull period for EM currencies between 2003 and 2008. When the Developed Market Crisis struck in 2008/09, inflation then declined sharply exactly at the same time that EM currencies crashed. As soon as EM currencies recovered in 2009/10 inflation began to rise and when EM currencies began their recent precipitous decline between 2011 and 2015, so did EM inflation. Inflation only stabilised in 2016 and currencies have promptly followed suit.
We also examined the relationship between currencies and inflation for individual countries by running simple regressions of nominal exchange rates on CPI inflation using the 15 EM countries in the index. The results are shown in the table below. Since nominal exchange rates are denoted in units of local currency per US dollar, it follows that a statistically significant positive coefficient supports the thesis of conventional FX pass-through, while a negative coefficient implies that weaker currencies are associated with lower inflation (in contradiction to the FX pass-through thesis).
The main observation from the table is that there is no consistent evidence to support the conventional FX pass-through thesis. Seven countries exhibit conventional FX pass-through, while eight countries show the opposite. The relationships between inflation and currencies are only statistically significant in half of the sample and within this subset exactly half — Brazil, Chile, Colombia and South Africa — exhibit signs of conventional FX pass-through, while the other half — Mexico, Romania, Hungary and Turkey — display the opposite. The relationships between currencies and inflation at the level of individual countries are therefore so random that they could just as well have been generated through random coin tosses.
Looking at the index level results, why does inflation tend to rise when EM currencies rally and vice versa? The most likely explanation is that inflation and currencies are jointly determined by a third variable: capital flows. If availability of foreign financing constrains economic activity at the margin, then growth and inflation will both respond negatively to capital outflows. Conversely, growth and inflation will pick up when money comes back to the asset class.
The evidence in support of severe capital constraints in EM is strong. EM countries now account for nearly 60 per cent of global GDP (in PPP-adjusted terms), but they account for less than 20 per cent of global fixed income. Only about one third of EM countries even have access to global capital markets. Moreover, capital flows within EMs are highly distorted, because most institutional investors closely follow indices, which only cover 9 per cent of EM fixed income.
The fact that conventional FX pass-through is a myth has important investment implications. Bouts of EM currency weakness typically unleash deflationary forces, which enable central banks to cut rates, so yields can decline. This means that local bonds are not the highly pro-cyclical instruments they are made out to be. Investors should therefore aim to have permanent allocations to local currency bonds while managing FX risks independently.
Jan Dehn is head of research at Ashmore, an emerging markets-focused asset manager.