Tuesday, June 29, 2010

The Importance of Business Cycle Studies in Emerging Markets

We found this interesting paper by Professor Carlos Bautista of the University of the Philippines about Business Cycle Research in Emerging Markets. A complicated topic, but the added value of careful analysis of economic fluctuations in Emerging Market Economies is probably huge. Bautista wrote the paper in 2007, but its outcomes are definitely still important. Most countries in the world are struggling with recovery after the Global Crisis and conversion patterns in the economy are exactly what business cycle studies are all about.

Using a dataset going back at least 20 years, Professor Bautista analyzed the economic cycles in 5 important Emerging Market economies, namely South Korea, Mexico, Malaysia, the Philippines and Chile. These countries were selected and are interesting for a couple of reasons: 

  1. The availability of a sufficiently long historical data set with quarterly information;
  2. The presence of at least one crisis: you cannot build business cycle models if you don't have a data set that covers all the various states you want to define and forecast.
  3. It is of course also nice that we are studying a sub-set of Emerging Markets that differ to quite some extent. Two are from Latin America and three are from Asia. Mexico and South Korea are 'larger' Emerging Economies from an investor's perspective, Malaysia is getting more attention recently (partly due to the growing interest in Islamic Finance) and Chile and the Philippines are still relatively 'small' from an investor perspective but interesting enough for different reasons. Chile on the one hand is relatively well-governed, with a large pension system supporting the financial infrastructure. The countries commodity richness (copper) makes it an interesting country for investors as well. The Philippines is a larger, albeit poorer Asian economy but one with large growth potential. Goldman Sachs classified it in its Next-11 group of human capital rich Emerging Markets.
During the last 20 years a lot of progress was made in business cycle research. Hamilton and Raj (2002) 'New Directions in Business Cycle Research and Financial Analysis', in Advances in Markov Switching Models (same authors as editors, published by Physica Verlag) provides a nice overview. Bautista's article contains interesting information about developments as well.

The essence of the business cycle research is on the one hand that economic cycles are non-linear and that they contain a trend and cycle component. The non-linearity makes it hard to forecast business cycles using 'standard' uni- or multi-variate statistical techniques that are based on linearity. Bautista explains which approaches on can use to capture non-linearity. 


Most business cycle research in Developed Nations classifies periods as either 'expansion' or 'recession'. This approach does make sense in mature economies that normally show low to moderate growth and hardly any rapid growth, with recessions being more of an exception than a regularly returning phenomenon. Bautista explains that things are quite different in Emerging Markets. He proposes a classification consisting of three states:

  1. Rapid Growth
  2. Moderate Growth
  3. Recession (i.e. Negative Growth)
Recession is not as rare as it is in more diversified, developed economies. On the other hand: neither is rapid growth. During the last 10 years we have seen nice proof of this. The period started with rapid growth in Emerging Markets, with economic growth rates in excess of 5 and sometimes even 10 percent per annum. Although most Emerging Markets were not directly involved in the sub-prime crisis, contagion and international market correlations in the Globalized world we live in induced a situation in which they seemed to suffer more from the Global Crisis during 2008-10 than quite a few Western, developed nations. However, when looking at forecasts for GDP, Current Account balances and analyze the rebounds at the stock market we see that recovery seems to be strong as well. To some extent this was related to the fact that amidst international panic following the unwinding of the Global Crisis international investors and corporations through their portfolio and foreign direct investments were creating difficulties for these nations. But just as fast as money was withdrawn by some, others are recognizing the opportunity to buy assets in Emerging Market economies at relatively low prices and this has resulted in stock market rebounds during the last 12 months.

In Table 1 to 5 we present data from the Bautista study in combination with output from the LMG GTAA model. As stated earlier, in the introduction, the Bautista paper was written in 2007 and we felt that it was interesting to see to what extent lessons could be learned from his study.

The 5 countries in the Bautista study can be categorized into three categories. Category 1 consists of South Korea, with a GDP per capita of USD 22,264 the richest of the five. Actually this level of GDP per capita resembles that of Developed countries. Recently MSCI reclassified Israel from the Emerging Markets group into the Developed group, but we wouldn't be surprised at all to see South Korea move from MSCI Emerging Markets to MSCI World within the next 12-24 months. Note: in our analyses we use the LMG GTAA calculation method for the GDP calculation, which implies that we take the equally-weighted average of nominal GDP per capita and GDP per capita corrected for purchasing power differences between countries.

Table 1 ; Selected data South Korea

The data on the left-hand side of the table are taken from the Bautista paper. In the rows we see the classification of individual quarters in 'Rapid Growth', 'Moderate Growth' and 'Recession'. Bautista does not rely on government institutions for the classification. Most Emerging countries do not have sophisticated, independent research institutions like the National Bureau of Economic Research in the US. Instead, he uses so-called Markov Switching models to classify quarters. For more detailed information about the approach, see his paper.

Table 4 in the Bautista paper shows per country what the probability of occurrence of a specific state of the economy in the next quarter is, based on the current state of the economy in this quarter. The left-hand side of our table lists these probabilities. Note that we have switched rows and columns compared to the Bautista paper for ease of presentation. According to Bautista it is more or less impossible for regimes to move in one go from 'rapid growth' to 'recession' or vice versa. This is incorporated in the table through probabilities of zero percent for cell 1,3 and 3,1 in the matrix component of the table. South Korea is clearly a moderate growth country, with a 61.2 percent probability of any quarter - without knowing what the previous quarter(s) were - being a 'moderate growth' quarter. The likelihood of 'rapid growth' is larger than that of 'recession', but a probability of 14.4 percent for a 'recession' is still quite large. Definitely not something smart investors can neglect.

South Korea has a relatively high market beta of 1.402 and a monthly stock market volatility of 11.91%, which is mainly indicative of the country being quite sensitive to international shocks. South Korea with its export-oriented economy benefits strongly when the effects of Globalization are positive, but is hurt quite substantially when Developed Nations struggle with the Global Crisis. The cumulative stock return over the last 3 years is still bad (-19.2 percent), albeit that the Seoul Stock Exchange did recover strongly during the last twelve months.
The second group of countries consists of Mexico, Malaysia and Chile. As you can see in Table 2, 3 and 4 all three have comparable GDP's per capita between USD 11,000 and USD 12,000.

Table 2 Selected data Mexico

Mexico is a 'moderate growth' economy, just like South Korea. Its Market Beta and Volatility are once again relatively large, indicating an above-average sensitivity to international priced risk factors. We are far less positive about expected GDP growth for the period 2010-2012 than we were for South Korea. This is based on the fact that Mexico is to a larger extent linked to the struggling US economy, whereas South Korea can benefit to a stronger extent from a rebound in the economic climate in Asia.
Table 3 Selected data Malaysia

When looking at the matrix component in the table, we see that the Malaysian economy differs quite a bit from the other two. This is not a 'moderate growth' economy but one that reports 'rapid growth' as the state of the economy with the highest probability of occurrence. Sounds great, but we should remember that it comes at a price: the likelihood of ending up in a 'recession' quarter is about twice as high as in South Korea and Mexico. To some extent this is related to the fact that Malaysia is a smaller, less-diversified economy which leads to larger sensitivities to external and domestic shocks in general. Second, Malaysia has seen good growth in the Financial Services industry. An industry that is normally not developed well in Emerging Markets, and therefore often a cause for the sensitivity of these countries to negative events. However, Malaysia has seen substantial growth in Islamic Finance during a period in which this form of Financial Services has become more popular. The exposure of Malaysia to what was going on in the Western world during the Global Crisis is lower than that of Mexico and South Korea. This translates into a relatively low stock market beta, and even volatility levels are not too high. Islamic Finance could have helped firms avoid the problems that became so typical for Western firms and consumers during the Global Crisis. Malaysia did therefore post a positive stock market return (cumulative over the last 3 years), albeit that returns over the last 12 months were more or less in line with those in the other countries.
The third country in the second category (middle zone) is Chile.

Table 4 Selected data Chile

Chile is a 'moderate growth' economy, just like South Korea and Mexico. The 'recession sensitivity' is lower than in the other economies. Public Debt is relatively low, economic growth looks reasonably good, the sensitivity to the rest of the world - as measured via its market beta - is not extreme and stock market performance was reasonable too. We believe that there are two explanations for Chile's good performance. On the one hand, the country's commodity base (Copper Mines) is very important with growing demands from Emerging, industrializing nations (e.g. China). On the other, Chile did create a pension system during the 60s and 70s of the previous century that is quite impressive by Emerging Markets standards. This system provides the economy with a kind of 'put option' device that helps avoid huge setbacks in times of economic hardship. The cumulative stock market return over the last 3 years was actually not so bad: 22.07 percent translate to about 7 percent per annum. A good return during a period of low interest rates. And during the last 12 months the 26.76 percent return is more or less in line with what the other Emerging Markets are doing.
Last but not least, Bautista's own country - the Philippines - can be considered a category by itself. This is the Emerging economy with - by far - the lowest GDP per capita.

Table 5 Selected data Philippines

The Philippines look quite a bit like the other Southeast Asian economy in our sample, Malaysia. The probability of a high growth quarter is relatively high, but so is the probability of a recession quarter. It is a complicated economy where all three states of the economy do have probabilities that do not differ that much when comparing them with each other. 'Moderate growth' is most likely but 'Rapid Growth' is almost as likely! The political situation is always complicated and although democracy has been reestablished after the Marcos years, it is clear that there is still a large 'inner circle' present with firm grip to power. In July 2010 the son of former president Cory Aquino and her assessed husband Benigno Aquino, Noynoy Aquino will become the next President of the country, replacing Gloria Arroyo. Starting from a lower base, economic growth is expected to reach levels similar to the best countries in our sample. The sensitivity to outside shocks is not extremely large, but the internal situation seems still less stable. We have reported earlier on the Philippines and refer to those entries.


What we can learn from the Bautista paper and the analysis above is the following:
  • There is no such thing as 'THE'  Emerging Markets. Although market movements in the financial market place might move in synch in the short-term - basically due to international money flows generated by international investors -, the long differences are clearly visible. This is already the case when comparing the five countries in the Bautista study, and even more so when analyzing all 30 countries in the LMG GTAA model.
  • Emerging Markets are for more sensitive to periods of economic hardship and the likelihood of recessions is larger. However, the potential for recovery is larger as well. Not a surprise, because these economies are less mature. This implies that once the tide has changed and bad periods are turned around and replaced by positive news, growth will recover quickly and periods of 'rapid growth' are more likely than in Developed Nations as well.
  • This translates into a plea for a business cycle approach that distinguishes between three states of the economy and not just two, like in many business cycle studies focusing on Developed Nations.
  • The linkage between macro-economic developments and developments in the financial markets is still relatively large in Emerging Markets where the importance of domestic factors is often larger - relatively speaking - than in Developed Nations where international market correlations are higher.
This implies that Bautista's study provides us with an important starting point for ongoing analysis of Emerging Markets. It would be interesting if a) the number of countries is increased so as to cover the whole MSCI Emerging Markets and MSCI Frontier Markets; and b) update forecasts for that spectrum on a regular (read: quarterly) basis. The more large institutional investors know about business cycle patterns in these countries, the larger the probability that they will act less panicky in periods of crisis. With long-term, structural economic growth prospects in these nations relatively larger than in Developed Nations this will help avoid unnecessary loss of returns and this will translate in a win-win for institutional investors in the Western world and firms / populations in Emerging Countries.

No comments:

Post a Comment