Global Asset Allocation: The Skater Does Not Travel in a Straight Line

Print ArticleEmail Article

by Victor A. Canto,  La Jolla Economics (Guest Contributor)

In an earlier paper we argued that in an unfettered market, arbitrage and the profit motive will insure that, in the long run, differences in rates of returns are equalized across different localities. It is apparent that transportation costs will prevent the complete equalization of prices and/or factor returns. Put another way, it will not be profitable to arbitrage differences in prices that exceed transportation or transaction costs. A more subtle insight is that the arbitrage process may take place through the commodity or asset with the lowest transaction costs. A somewhat related issue is the convergence process. It seems reasonable to assume that the fastest the adjustment to a new equilibrium, the higher the adjustment costs and, thus, the greater the short run “transportation costs”.

Arbitrage and the Strategic Global Allocation

The data suggests that the optimal adjustment process for both GDP and market capitalization may be slower than what is technically feasible at any point in time. While there may be adjustment costs slowing down the convergence process, over the long run these “transportation costs” will be miniscule. Thus, the convergence will be almost complete. So the question is, how do we take advantage of the long run convergence?

We then used an insight from Wayne Gretzky. When asked why he was so good, he simply said that he skates to where the puck is going to be. Gretzky’s insight may have some value to a global asset allocation process. If you know where the puck is going to be, why not skate there. In context of a global asset allocation decision making process, how much to allocate to the international component of a portfolio. The way we approach the issue is fairly straightforward. The first question we tend to ask is, if a country’s share of world GDP is 25% what should that country’s share of the world market capitalization be? The most frequent answer that we get is 25% and that makes a lot of sense. It suggests that in the long run the ratio of a country’s share of the world wealth to the country’s share of the world GDP will approach 1. A more practical way to say the same thing is that arbitrage will insure that valuations will converge across the world. We use the ratio of the market capitalization of regions to the regions’ GDP in Figure 1.

Although there are differences in the levels, Figure 1 shows that there is a close correlation between the world’s valuation and that of the U.S. The close correlation, with its implication that in the long run market valuations are going to converge, gives us a simple insight as to the optimal long run/strategic global allocation. If the evidence presented in the earlier paper is of any value, it suggests that in a bullish scenario the faster growing and lower valuation regions of the world, such as emerging markets, are going to appreciate faster than the developed world. Likewise, in a bearish scenario, emerging markets may depreciate less. If so why not allocate to where the market is going to be rather than rebalance to a benchmark that is always lagging the true global allocation? Which adjustment process an investor selects could very well be the difference between a superior performance and a lagging performance.

“Transportation Costs” and the Cyclical Global Allocation

The data presented in Figure 1 shows that the valuation levels are not identical and that the correlations are not perfect. First, there is a difference in levels. A second difference is also visually discernible. There are some periods where the two valuation series deviate from each other.

Our explanation for the cycles is that government policies can create “transportation and adjustment costs” that would prevent the complete equalization of valuation and, thus, the differences in levels shown in Figure 1. Also, changes in the “transportation costs” could either accelerate and/or retard the long run valuation convergence process. In fact, in some cases the policy actions may even temporarily reverse the natural equilibrating process. Thus, the adjustment to a long run equilibrium valuation measure is not a straight line adjustment. Getting back to the Gretzky quote, while it is true that the puck travels in a straight line, the skater does not, nor may it be optimal to, skate in a straight line to be where the puck is gong to be.

Policy changes go a long way to explain much of the differences in behavior among the regional valuations series. Figure 2 shows the USA valuation proxy as measured by the ratio of market capitalization to GDP. We also included the various presidential periods in order to see if there is any correlation between the different administrations and the valuation measure. The data suggests that there is, but there is more to the story. The policy mix matters a great deal. The common thread between Nixon and Carter was rising inflation, high tax rates and an increase in the regulatory burden. Reagan changed all that and adopted a deregulation, lower tax rate policy combined with a tight money (i.e. domestic price rule) policy. The policy mix ushered in an economic recovery and a rise in market valuation. Papa Bush “read my lips” is the only hiccup in the data. However, as seen in Figure 2, the valuation flattened during the second half of the Bush administration and continued during Clinton. In fact, the little dip is associated with the Clinton tax rate increase. However, once gridlock took effect and the Maestro found his “magic wand”, valuation shot up. The economy prospered and a bubble may have been created when the Maestro eased too much in anticipation of Y2K and then removed the punch bowl when Y2K did not have an adverse effect on the economy.

Bush was elected and the economy did not take off until the tax rate was cut once again. However, the Fed’s accommodative policies may have ushered in another bubble bout. This time the economy had a crash landing and went into a deep recession. President Bush tried to stimulate the economy with a tax cut that “put money in people’s pocket”. It did not work. The economy’s valuation did not surge until capital gains and dividend tax rates were lowered to 15%.

Figure 3 tells pretty much the same story. During the stagflation years the U.S. lost market share in both GDP and market capitalization. The election of Ronald Reagan marked a turning point and the U.S. experienced a secular increase in their share of both market capitalization and GDP. The secular increase peaked around 2000. After that the U.S. began a market share losing trend. If we can anticipate these temporary deviations, we may be able to take advantage of these cycles and develop a cyclical global allocation strategy, which we call the LJE ValueTimingTM strategy.

The insight of free mobility and free trade implies mobile factor returns will be equalized across regions. Hence, the supply response of these factors will impart a common trend among world economies. The differences in performance will emanate from the responses of the immobile factors across regions. For example, if the positive economic policy shocks emanate in the U.S., we expect to observe a rise in economic activity in the U.S. as factor returns rise. The price signal will lead to an increase in the supply of the mobile goods in the rest of the world. Valuation and income will rise in the rest of the world. However, to the extent that the positive shocks are localized, it will have no impact on the valuation of foreign immobile factors. Hence, we should see U.S. output outperform the rest of the world. However, if the rest of the world reacts to the U.S. policies, then it is possible to see a surge in their valuation. That is exactly what happened to the rest of the world shortly after the Reagan tax rate cuts. Many other countries followed suit and they experienced a surge in valuation (Figure 4). However, as time went on policies diverged a bit. The Europeans adopted the Euro with its mandate of 2% inflation while the U.S. was led by the Greenspan wand. Fiscal differences were also evident. We had gridlock and a reduction of the capital gains tax under the Clinton administration and then a reduction in the dividend and capital gains tax rate under Bush. These differences in policies led to a temporary divergence in valuation (Figure 4). In the Far East the differences were even more pronounced (Figure 5). Notice the abrupt decline in valuation around the end of the 1980s. It was at this time that Japan enacted its first capital gains tax. It marks Japan’s peak in valuation and it has taken Japan a long time since then to catch up to the rest of the world’s valuation levels again.

Convergence and Policy Cycles: Some Investment Insights

Figure 3 shows a strong correlation between the U.S.’s share of world income and its share of the world market cap. The data also shows that the correlation between the share of market valuation and the regions’ share of world GDP is not perfect. However, over the long term, the two will tend to move together. The question is how is the convergence achieved? Does the market capitalization adjust to the countries’ GDP growth or vice versa? We believe, that in the long run, GDP will rule the day. Now the investment question is whether one should adjust a global portfolio country allocation in anticipation of the coming changes? We think so.

In addition to the correlation between a country’s share of the world GDP and world market capitalization, we can also observe relative valuation cycles (Figure 6a and b). We are quite interested in the episodes where valuation divergences occur. If they are as we suspect, the cycles will be mean reverting. Under these circumstances it may be desirable to adjust the country weighting to take advantage of the temporary deviation of relative valuations. For example, in both Figure 6a and 6b the relative valuation inflection points are associated with relative changes in economic policies. The U.S. underperformed until the Reagan policy mix was put in place. A new inflection point occurred in 2000 when the rest of the world began to gain on us.

 Looking at the policy mix now being implemented in the U.S., our outlook for the U.S. is that it will underperform the world. Add to this the fact that the U.S. has a disproportionate share of the world’s market capitalization relative to its GDP and the secular trend is also for the U.S. to lose market share. Hence, on both accounts a reallocation of assets away from the U.S. to the rest of the world may be warranted.

Victor A. Canto
La Jolla Economics
www.lajollaeconomics.com