THE FOLLY OF PARTIAL EQUILIBRIUM THINKING: FALLING CENTRAL BANK FX RESERVE GROWTH IS NOT = RISING TREASURY YIELDS

Measuring the growth of global foreign exchange reserves was a popular macro-analyst pass-time during the pre-crisis low vol hey-days. This exersize was conducted to gauge the central demand for U.S. Treasuries. The simple conclusion from this analysis was that slowing official asset accumulation would result in declining demand and rising Treasury yields. This argument misses one crucial point.

The rapid ascension of official reserve assets was a result of the US current account deficit. The only scenario in which this deficit would have corrected was some combination of a rising US savings rate and falling investment. Thus, in the event of a sharp contraction of the trade gap, net issuance of various kinds of securities would have fallen as well. Clearly, in such a scenario, there would have been no pressure on yields to rise.

A cursory look at the US flow of funds data tells us that net issuance of various fixed income products is down sharply and the rising household savings are finding their way into the treasury market. Consequently, there seems to be no pressure on yields to rise even when foreign exchange reserve growth has slowed down (even in the face of mammoth issuance).

BUYING GROWTH STOCKS WITH A SAFETY VALVE

For the most part, I do not have any investing bent (macro, quantitative, value, growth, momentum etc) because I feel that the market is just too dynamic and no philosophy really works all the time. Even so, I often find it hard to convince myself to buy growth stocks selling at high multiples. The problem is that one negative growth surprise and the consequent multiple contraction has the potential to destroy performance.

There are a lot of companies I really like, but the market adores them too; leading to pricey valuations. One such stock is Jain Irrigation. India has numerous and growing mouths to feed. The erratic monsoon this year has again reminded the nation that agriculture continues to be largely rain-fed. Micro-irrigation is the solution and Jain Irrigation is the leader that space. The market acknowledges this fact and rewards the company with premium multiples. That is where my problem starts. My solution to this quandary is to figure out a way to get out with minimal/controlled draw down if growth disappoints in the future.

To guard against precipitous downside, I tested a simple technical rule. I would own the stock only when it is trading above the 20-day moving average. The results (posted below) are somewhat encouraging.


Jain Irrigation Yearly Return Summary Statistics: Jan 1991 to Dec 2008



Jain Irrigation: Jan 1991 to Dec 2008 (Jan 1991 = 100)



Source: Bloomberg

INDIAN EQUITIES STILL PRICED RIGHT FOR MEDIUM-TERM INVESTORS

IN MY VIEW, 10-YEAR COMPOUNDED ANNUAL REAL RETURNS SHOULD BE ABOUT 8% WITH AN UPWARD BIAS

On December 7, 2008, I held the opinion that Indian equities were attractively priced for investors with a medium-term horizon (SENSEX PRICED TO DELIVER ATTRACTIVE MEDIUM-TERM RETURNS). I point that out not to stake my claim for predicting the breathtaking Q2 2009 rally, but to highlight that buying at the right price is key to investment success.

Short-term returns are driven by ebbs and flow in risk appetite. Such moves can be hard to predict even for the most seasoned market watchers. Over longer-term however, valuation plays the dominant role in determining returns. Sharp valuation downdrafts are good entry points for investors willing to bear near-term volatility.

EQUITIES STILL PRICED TO BEAT BONDS OVER 5 to 10-YEARS

In my previous piece, I had valued the SENSEX based on a price to nominal GDP multiple. Viewed from the same lense, equities are no longer under-valued. Yet, the asset class is not exuberantly priced either. From current levels, investors could garner at least 8% real returns over the next 10-years. This forecast assumes mean historical valuations and real GDP growth of about 7% to 8%. Needless to say that this projection builds in a margin of safety.

Yet, we must invest fully cognizant of the fact that over the short-term there could still be another downdraft.

FINANCIALLY ENGINEERING AN "ALL-WEATHER" EQUITY PORTFOLIO

In principle, any asset class or security can be levered up/down to create the desired risk return payoff. Based on this concept, two negatively correlated securities can be combined to achieve a given level of expected return with lower risk (defined in terms of standard deviation). Using this paradigm, I seek to configure a portfolio with the expected return of equities, but lower standard deviation: the "all weather equity portfolio".

Treasury yields/prices and equities are positively/negatively correlated (see exhibit 1). This property is particularly helpful in creating the "all the weather portfolio". During the 20th century, U.S. stocks and bonds generated yearly real returns of 6.7% and 1.6%, respectively. Assuming 2% inflation going forward, I construct the portfolio as follows:

1. Long U.S. Treasuries: 1.25
2. S&P 500: 0.5
3. Borrowing: -0.75

Effectively, I buy U.S. Treasuries on leverage to achieve equity like returns. I used ETFs to conduct this exercise for the period July 30, 2002 to June 11, 2009: TLT: + 20yr Treasuries and SPY. I assume leverage at 8.5% (the margin rate for a personal investor). Institutional investors can finance the bond portfolio cheaply in the GC-repo market. Finally, the portfolio is rebalanced every day.

The performance of the portfolio can be observed in exhibit 2. Importantly, while the S&P 500 rose about 5% during the period under consideration, the all weather portfolio was up 11%. At the peak of the price appreciation cycle, stocks were up about 74%, while our financially engineered portfolio climbed only 42%. This under performance in the bull market is a function of the high interest rate I used for assuming leverage. Interestingly, the all-weather portfolio reached its peak with 62% total return in December 2008 – a time when the equity only portfolio had failed to generate any return since inception.

Redoing the calculation with the GC-repo rate (average for the period: 2.54%), I get more favorable (and more realistic for institutional investors) results (see exhibit 2). In this case, the all weather portfolio keeps pace with equities in the up move, and protects downside during the market crash of 2008.


Exhibit 1. Positive Correlation: U.S. Equities vs. U.S. Treasury Yields



Exhibit 2. "All-Weather" Portfolio Performance: Institutional



Exhibit 3. "All-Weather" Portfolio Performance: Personal



Exhibit 4. Daily Return Statistics




Geek Notes

For calculating the total returns of the all-weather portfolio, I used the average yield on the 30-year U.S. Treasury as a proxy for coupon. And, I credited this coupon return on a daily basis. This implies that the returns on the all-weather portfolio are somewhat over-stated due to daily compounding.

Source: Bloomberg
Dimson, Staunton and Marsh; Triumph of the Optimists
Bridgewater Associates

PERILS OF OVER-EMPHASIZING A FEW TURNING POINTS

GDS: Gross Domestic Savings (Fiscal Year)
GDCF: Gross Domestic Capital Formation(Fiscal Year)
Sensex year-end (Calendar Year)
Source: Bloomberg; RBI

It seems from this graph that the Indian equity market delivers sustained stellar returns when the gross domestic savings and capital formation rise significantly. This suggests that superior market moves coincide with periods of rising trend growth rate of the economy (for the logic please see my previous post). Careful statistical analysis, however, points to a positive but statistically tenuous relationship.


SENSEX RETURN = 20.90(0.0034) + 0.93*GDFC Y-o-Y%(0.2645)
P-values in the brackets.

R^2= 5%

The obvious conclusion is that economic growth is only one part of the puzzle. Valuation is important. Rapid growth which has already been discounted by the market will not help.

9% GROWTH IN THE FACE OF A GLOBAL RECESSION: DREAM ON; ANY EQUITY RALLY BUILT ON THIS THEME WILL FACE GROWTH DISSAPOINTMENT IN THE FUTURE

The India growth story is a strong one. India has the most important advantage: top quality human capital. Not many countries with comparable social parameters can boast of India’s success in some quarters of higher education.

My view of Indian Trend Growth Dynamics: With Infinite Supply of Labor on the Margin, Trend Growth Rate is Determined by Productive Capacity: Gross Domestic Capital Formation and Ultimately on Funding Sources = Gross Domestic Savings + Sustainable Level of Current Account Deficit

1. Exports pick up leading to an increase in capacity utlization. A surge in capital formation results, which increases the productive capacity of the economy.

2. Initially, the current account deficit widens as funding needs outpace domestic savings. Ultimately, savings pick up as the propensity to save is higher from profits than from wages. And also because at higher levels of income, there is more leg room to save.

3. The higher gross savings rate leads to a rise in trend growth in terms of sustainability of funding available to finance the higher level of capital expenditure.

4. The down-leg begins with a down-turn in exports which spills-over into capex through the capacity utlization route.

5. Consumption spending is stable and grows at the new trend rate, providing stability to the economic ascension.

6. The savings rate dips in the downturn.

Rising trend growth is the norm for the Indian economy. Yet, do not underestimate the importance of cyclical forces.

Figure 1. Since the Mid-80s, cycles in GDP growth have been closely related to perturbations in exports. I have calculated the cyclical component by treating a 5-year moving average of GDP growth as the trend.



Figure 2. Cycles in Gross domestic capital formation are intricately linked to external sector performance.



Figure 3. Private final consumption expenditure lends stability to the expansion.



Figure 4. Government final consumption expenditure runs counter to the cycle, implying a countercyclical fiscal policy.





BOTTOM LINE

India’s trend growth has risen in the last 5-years to around 7%-8% as the gross domestic savings rate has climbed to about 32% (assuming an ICOR of 4). In the next big capex cycle, the trend will rise to 9%-10%. But that is not likely to happen without a recovery in global demand. And we should not forget that the last 5-years were more an exception than the norm. Until then, 6%-7% is more likely to be the outcome. That's pretty good, but it is way below 9%.


Source: Handbook of Statistics, RBI
All data at constant prices.
GFCE: Government final consumption expenditure.
PFCE: Private final consumption expenditure.
GDCF: Gross domestic capital formation.

VALUATION MATTERS: RISK-ADJUSTED RETURN PERFORMANCE: U.S. HIGH YIELD TRUMPS S&P 500; USING LEVERAGE TO ENGINEER SUPERIOR RETURNS GIVEN RISK CONSTRAINT

The S&P 500 has rallied almost 40% from the March 9, 2009 lows. U.S. HY (HYG) and U.S. IG (LQD) are up 28% and 8%, respectively.

Yet, risk adjusted performance tells a different story. And that is what really matters because any asset class/security can be levered up/down to provide the desired risk-return characteristics. Clearly, the investment with the highest ratio of mean to standard deviation would dominate in this framework.

Based on this criteria(see Figure 1), U.S. High Yield performed the best in the rally since March 9, 2009, followed by equities and U.S. IG.

Assuming that we could borrow at 10%(see Figure 1), it was possible to engineer a portfolio that would have outperformed stocks while commanding the same risk defined in terms of standard deviation.

The key message of the story is that valuation matters. At the March 9 lows, stocks were modestly undervalued, but U.S. High-Yield looked battered and offered better potential risk-reward characteristics.

Figure 1