A simple yet effective EMFX trend and valuation tool

Remember, EMFX is essentially one big beta play

One relationship I came across was between EM and DM exports, and the changes in EMFX. This was related to a piece written by Jonathan Anderson (“How to Think About Emerging Markets (Part 10) – Macro and EM Currencies”) where he illustrated how closely EMFX performance was related to global growth – proxied by EM exports. By his calculations, the correlation was between 80–85% for the past 20 years.

Taking a slightly different angle on this, we can see the ratio of EM to G3 Exports (YoY% change) to the YoY% change in the JPM ELMI+ Index is quite tight, as per below. Note, for my ‘EM’ universe of exports I have used only China, Korea and Chile. I believe these 3 are the most important and timely. G3 is US, EU and Japan exports.

Green line – EM to G3 Ratio of YoY% change in Exports

Pink line – YoY% change in JPM ELMI+ Index

The correlation between these two series is 73%, with an R-Squared of 0.54.

So now we have a fundamental trend determinant in our tool kit.

A natural follow-on comes from this relationship, and that is to construct a ‘predicted’ EMFX index based upon a regression of the series. Doing this will enable us to use the ‘predicted’ index as a barometer of over- or undervaluation of EMFX. This looks as follows:

Red line – ‘Predicted’ JPM ELMI+ Index (YoY% change)

Blue line – Actual JPM ELMI+ Index (YoY% change)

Black lines – 1 Std Dev bands of ‘predicted’ line

Putting 1 Std Dev bands on the ‘Predicted’ line gives us a sense of over- or undervaluation relative to where the actual index ‘should’ be trading.

Current conclusions on this framework: the trend of EMFX might have turned recently as the export ratio has had a turn down – although possibly too early to conclude decisively, and secondly EMFX is currently overvalued.