Monday, September 30, 2019

Endowment performance and smart money - There will be a need for liquidity

Many of the largest and best university endowment have reported their returns for the first half of the year. The numbers are all positive, but that does not mean these endowments have beat simple benchmarks. For the same period, a simple 60/40 SPY/AGG combination would have returned 12.95 percent return which is just above the best endowment managers in our limited sample. Simple can be better and getting the asset class selection right is still the driver of performance.

There was surprising return dispersion with the lowest manager, Yale University, generating half the return of the best, Brown University. The return dispersion for all of 2018 was only slightly larger. When volatility increases, so will return dispersion. The investment choices of managers, good or bad, become more obvious when return volatility and correlation changes.

The long-run strategy of Yale does not mean that it will outperform peers in all short-run periods. Yale was above the median in 2018. Any short-run ranking can bounce from best to worst in the short; however, it is critical to catch the absolute turning points in market cycles. 

Our greatest fear is that the move to private equity through the endowment model over the last decade will leave many university portfolios stuck with illiquid positions. Endowments do have the advantage of a long-run horizon, but there is a tipping point where liquidity becomes more valuable than high returns. As we come closer to any market turn, liquid strategies increase in value.

Tuesday, September 17, 2019

Are Alternative Risk Premia Effective Hedge Fund Replacements?

Are alternative risk premia (ARP) portfolios a good replacement for hedge funds? A recent article "Oft-Touted Risk Premia Strategies Show Their Weak Side" commented on the lagging performance of risk premia over the last few quarters and questioned whether ARPs have lived up to expectations as a hedge fund replacement. Hedge funds have fared well in 2019, but many risk premia strategies have also shown a pick-up in performance this year. Similarly, some ARPs have correlated with poorer performance with some focused hedge fund styles.

The real comparison is not between a single ARP and a hedge fund style but a bundle of ARPs and a hedge funds. Hedge funds are usually not monolithic strategies but rather a combination of different strategies weighted by risk exposures. There will be hedge funds that specialize with a single strategy or with the majority of risk exposure in single strategy but most will be described by more than one factor.

There will be performance differences between ARP portfolios executed through bank swaps and hedge fund returns because they are fundamentally different views on how portfolios and investments should be structured and managed. These distinctions are important whether you want to use ARP as complement or as a substitute for hedge funds. 

ARPs should be viewed as factor building blocks that can be bundled into portfolios with well-defined risks. Hedge funds are managed investments that have factor risk exposures that can be categorized through alternative risk premia. One is a direct expression of risk factors while the other is a manager's representation of risk factors with the potential addition of skill. In the case of hedge funds, these representations may do better or worse than a specific weighted set of factors. The relative outperformance may be skill or may be caused by a poor measure of the risk taken. ARPs, on the other had can be bundled in any combination desired; however, the underlying ARP swaps are restricted by the rules used for construction.

Our table provides some of the distinctions between an ARP swap portfolios and hedge funds and offers insights on which investment choice has an advantage. An investor can run through this list and determine their preferences.   

We view that ARPs can be compared with hedge funds on a number of different dimensions: strategy, portfolio, flexibility, alpha generation, cost, transparency, and liquidity. In most cases ARP structures will have an advantage over hedge funds. The key difference is manager skill, yet in many cases, when properly measured, alpha is limited. Hedge fund managers may have a slight advantage with the dynamic adjustment of strategy exposure, yet even in this case there are systematic and inexpensive ways to adjust exposures that can give ARP an edge. 

Alternative risk premia will have dynamic returns which will not always be positive, but when properly compared with hedge funds, ARP will have clearly measured advantages.

Negativzin is a strafzin and investors are not happy with rates

The ECB has a solution for economic growth in Europe, more pain for savers. Germans don't refer to rates as negative but as a punishment, "straf". You will be punished until you are willing to consumer more and stop this silly idea of savings for future uncertainty. 

As a final act, president Draghi reopened their QE program and lowered rates to inflict more savings pain. Of course, in Rube Goldberg fashion, the ECB idea is to lower rates but actually not have the impact felt too much by depositors. This is a policy of selective pain. Can the policy-makers nudge the banks to lend without hurting the banks through crushing the earnings of financial institutions.

This rate policy does not change government spending behavior so there is no coordination between monetary and fiscal policy. There is no end is sight for negative rates. These expectations means that investors haver to either buy riskier assets for yield or save more not less. Capital flows will move to those places that have higher relative rates. Negative rates will be exported around the world with no solution 

Monday, September 16, 2019

Momentum meltdown not a trend-following meltdown

Trend-following is not momentum investing. They are similar and have many of the same characteristics and have been often described as being the same, but investors should not be confused. There has been a momentum meltdown this month, but there has not been a trend-following meltdown. 

Trend-following looks at times series behavior for a wide range of markets and will buy (sell) those that are moving higher (lower) based on some absolute criteria. There is not the expectation that the long trends will be matched against short trends Momentum strategies often applied to equity markets are cross-sectional. Market momentum is measured and ranked with the portfolio going long the highest momentum stocks and short the lowest momentum. The portfolio will be a long/short match based on these relative ranking.   

There has been a positive change in the macro environment which has hurt the strong bond trends seen since spring. This increase in bond yields has shocked many trend-followers who have been long the global bonds, but this move is independent of the momentum crash in equities which have seen a rotation from momentum growth to value stocks. There may be similar causes but the factor styles are different. 

Trend-followers generally focus on futures markets and not individual equities, hence the reference to managed futures. There can be further revisions from  changes in macro markets but this is not the same as a style factor rotation in equities.

Using CSAIX as a managed futures trend-following proxy and MOM as momentum proxy, we can see that they are related but the trend-following is more closely related to bond positioning and behavior. The trend-following program is less likely to have crash effects given its diversification. During this same time, there has been a positive shock to the value style. Herding behavior is often captured in the momentum strategy and will be subject to crashes when there is a major change in expectations. Trend-following may also be subject to trend revisions but the drivers will be associated with broader market changes.

Saturday, September 14, 2019

The end of Chimerica and the end of world order liberalism

The 21st century has been the era of Chimerica. The symbiotic relationship between China and US drove global growth and brought these two countries (two systems) into a close relationship of trade and finance. In 2018, China was the number three goods export market for the US and the number one goods importer. The two countries had overall goods and services trade exceeding $730 billion and a trade deficit of just under $400 billion. The US is running a trade and savings deficit with goods coming to America and dollars building up in China.

These trade numbers have not been higher, yet philosophically there is a now a sea change in thinking about link between the two countries. There has been the belief that trade and cooperation would lead to change in China with deeper cooperation instead of rivalry like superpowers in the 20th century.

The reality and dream of Chimerica was not perfect and was not all together healthy, but it was a trade and financial flow reality. Niall Ferguson, the historian, coined the phase a little over decade ago as an apt description of the times, but the Chimerica of the past is done regardless of any trade deal or the actual trade numbers. (See "Make Chimerica Great Again" by Ferguesen and Xu of the Hoover Institute to get their view on Chimerica.) There is now a political skepticism between the two countries coupled with super-power competition that is unlikely to reverse to anything we saw even three years ago. Trade may continue but in a wary form different from classic libertarian trade. China is now a strategic competitor not a strategic partner. 

There may not be a reversal of financial deals but the world has to find a new framework because the dual power structure of west and east is not sustainable. Perhaps there is a new darker Chimerica. The framework of looking at China as another large economy for investing is in flux regardless of index inclusion or market capitalization. A world of rivalry is less efficient with trade and capital flow driven by new criteria independent of comparative advantage or return on investment. 

Idea, culture, and people flows will dwindle in this darker world. Cooperative optimism has turned into mercantilistic pessimism. There will a new search for global partners in other countries. There is no room for a new order of liberalism for all countries. This will have a further spillover for all emerging market investing with greater instability and uncertainty. It is a headwind that cannot be ignored.

Friday, September 13, 2019

Active open-mindedness - Eliminating some behavioral biases

How do you become a better investment analyst? Simple, have an active open mind. Active open-mindedness or as some have described, active open-minded thinking (AOT) can be boiled down to developing two important skills. One, the open-minded thinker learns to use all new information that is available. An active open minded thinker will be open to new and possibly conflicting information. This type of thinker increases the level of search for information without the demand for direction or confirmation. Two, the active open-minded thinker receives new information, learns, and changes his mind when presented with new facts. He is open to changing views when presented with new information. Some philosophers have called this a "reflective equilibrium".

An AOT process can help reduce some of the key behavioral biases that affect investors. There are biases that can be categorized by their misuse of information and whether the decision-making is open to change. For example, having a confirmation bias would be the opposite of open-mindedness. AOT wants to end any "my-side bias" associated the use of information and inference. Investors should move from my-side to contemplating other-side arguments. There have been tests devised to tell whether someone is open-mined. AOT can be measured and the level of openness can be enhanced.

AOT is more than just being open-minded but actively seeking alternative thinking. It means being more reflective and less impulsive where the main impulse is to maintain the status quo opinion. This could be viewed as a corollary to Kahneman's fast versus slow thinking that is the basis for heuristics and behavioral biases. 

This process of open-mindedness is not an easy skill to acquire, yet it is critical even for those that are systematic model-builders. Start slow; just ask the simple question, "What information do I need to change my opinion?"

Tuesday, September 10, 2019

Policy Vandalism And The Savings Glut - Negative Interest Rates Are Not Going to Reverse

Negative interest rates are a financial existential threat to the global economy. Odd investment decision and behavior flow from the continuation of negative rates. All savers and borrowers are affected. Every pension is impacted. Every set of cash flows that has to be discounted is distorted. Negative rates have moved from a temporary event associated with a financial crisis to the norm for most economies. 

While negative nominal rates were an oddity until the 21st century, the focus also has to be on the real rate of interest. The real rate has also been low for an extended period. Low expected inflation coupled with low or negative nominal rates creates low or negative real rates. Low real rates used to be a temporary event associated with surprise inflation. It is now the norm in the 21st century.

Central bank policies that further push short nominal rates negative for long time periods with regulatory pressure is a form of policy vandalism. Negative rates are a distortion that carries over to other financial assets and distorts investor behavior.  However, these policies are linked to the systemic ongoing savings glut around the world. 

Savings has to match investments and any mismatch will have to be adjusted through price, the rate of interest. As measured through a number of models, the neutral rate or r-star has fallen close to zero or through the zero bound. This is a long-term decline. Central banks will say their hands are tied to what they can do. They are only responding to low neutral rates. Policy rates have fallen as a response to these secular declines in equilibrium rates. They argue that they cannot be vandals because the saving glut made them force rates negative. 

The savings glut argument is hard for investors to fully appreciate and exploit because it is a long-term phenomena and not a short-term trade. The combination of demographics and technology has changed the balance between savings and investment. Demographics do not just represent a slowing of population and economic growth. Current aging demographics create an increasing divide between retirement and death that has changed savings especially in advanced economies. Age bubbles play havoc on country savings rates. The time value of money may break its normal relationship to actually have savings place more value on future consumption than on current consumption. 

As important, there is a shortage of investments. For EM, there are limited safe assets. In developed economies, there has been an embracing of austerity which limits fiscal policy as an investment driver. Technology also has changed the needs for capital. A service and information based economy just does not need the same requirements as an industrial based economy. This only exacerbates the savings investment imbalance.

Central banks cannot end the savings imbalance. They just allow a continuation of the problem and further accommodate the ongoing secular rate decline. They can make it worse by asserting that they have control of the process. Their illusion of control actually adds to uncertainty. 

Investors need to accept that rates, real or nominal, are not moving higher under the status quo. The hard planning based on a future that will not go back to anything like the pre-Financial Crisis is necessary.

Monday, September 9, 2019

Alternative weighting schemes and smart beta - Comparing Sharpe ratios

Can investors improve upon a cap-weighted equity benchmark? This is a fundamental question for many investors. It is also one of the foundational issues for smart beta products that use different passive weighing schemes. Hat tip to Adam Butler of @GestaltU for referencing this piece of older research by Nick Motson of the Cass Business School in London. Motson did exhaustive work on the issue and produced some compelling conclusions. (See Smart beta, Scrabble, and Simian Indices)

There are strong theoretical reasons for alternative weighing scheme but for most investors it is an empirical issue on whether it works over a long-run period with different return environments. It is hard for any investor to change benchmarks and of course any new benchmark has to be compared to the old standard, but the Motson work provides some interesting food for thought. The smart beta alternatives generally have higher Sharpe ratios, lower volatility, and alpha versus a market-weighted benchmark although there is no alpha versus a Fama-French framework. Alternative schemes can have high correlation with a classic cap-weighted index, but there can still be significant tracking error and return dispersion over short time periods. 

Of course, all of these alternative indices have one thing in common; they reduce the exposure to the largest cap stock names. Still, for investors who want some simple ways to systematically change equity risk exposure, the smart beta benchmarks are alternatives worth exploring. Note that these products will have higher fees and more transaction costs, so there is a question on how to effectively implement or structure portfolios to generate the intended result. It can be as simple as adding more exposure to lower market cap names.

Saturday, September 7, 2019

Hick's Law Applied to Investments - Too Many Darn Choices!

There is rule used to describe the relationship between choices and reaction time called the Hick-Hyman Law or Hick's Law. An increase in the number of choices presented for a decision leads to a longer reaction time. It has been applied to marketing, product development, education and other assorted fields. The law has not been applied to investments, yet the extension seems natural.

If there are more investment choices, whether mutual funds or ETFs, there will be a longer reaction time before any choice is made. If there are too many choices, then at the extreme no decision is made. 

More importantly, for traders, if there is more information to choose from, there will be a lower decision reaction time. If there are more policy choices for what action central bankers will take, there will be a longer reaction time.

There are solutions to the Hick's Law problem.  
  • Use models to cut down the choices and speed-up reaction. In the simplest case, the choice could be based on trends in price. A more complex model choice would be to use only a limited number of fundamental information inputs. A model will cut the processing time for any decision.
  • Follow rules. This is a variation on the model idea and will increase reaction time regardless of information used. 
  • Categorize information that is similar to cut the number of choices. For example, valuation measures can be bundled or aggregated for a single decision.
These solutions come in two types: one, reducing the choice set and cutting the noise and two, increasing the reaction time through hard-wiring how a decision will be made. By thinking about the basics of Hick's Law investors can fight the current problems of increased uncertainty from too many choices.

Thursday, September 5, 2019

Big managed futures returns - There is crisis alpha just not associated with equities

Based on the continued bond rally, August posted another good return month for managed futures (trend-followers). The major trend-following indices are now well into double-digit returns. There are classic trend-followers who are returning well above 20 percent, but there has not really been a "crisis" for this alpha generation as defined by the "crisis alpha" crowd. The whole crisis alpha story is based on large trend-following gains during periods of high equity stress. This crisis story is an ex post analysis. Investors don't know there is a crisis period until after the fact. The crisis is based on a measured decline (down 20 percent) in equity index returns. All large equity declines see strong trend-following gains, but that does not mean that all trend-following gains are only associated equity declines. Some major market "crises" have occurred in 2019. They just have not been in equities. 

We like to use the language of market divergence explaining trend-following gains. Market divergences or dislocations away from equilibrium or existing price ranges lead to trend-following gains. The return gains are a function of the size of move and fund exposure. There have been significant market divergences in fixed income, energy, and currency markets over the last. These have been more significant than what we have seen in the last three years. US Treasury bond futures have gained over 15 percent this year, oil has fallen 25 percent in the last year, gold is up over 25 percent in 2019, and selected currencies have moved close to 10 percent in the last few months. These examples are not even accounting for more localized markets distortions. All these markets have signaled divergences. Still, their impact have not yet fully spilled-over to equity markets. Equities seem to be discounting a different set of facts or a very different time line for risk events. 

There is a key feature of trend-following that make it attractive, yet it seems there is trend-following strategy fear inconsistent with simple utility maximization. Managed futures trend-followers have return distributions that show positive skew. There is a higher likelihood of extreme positive returns which is a characteristic that should be desired by investors. Investors like lottery bets and they should especially like lottery bets that pay-off during bad times. Managed futures gives positive skew for two simple reasons, holding trends and cutting loses. 

A trend-follower will hold trades that move away from some past equilibrium price. There is no attempt at finding a price target or valuation. It does not matter if there is a price increase or decrease. Trend-followers exploit dislocations and do better as market prices move to extremes. Trend-followers are described as holding long straddles and are long convexity. Regardless of the description, they will have more losing trades under stable prices with large winners when prices move to extremes. Given that trend-followers are non-predictive, it cannot be said when these large return events will come or when they may reverse. High returns cannot last forever, but given the ability to go both long and short means that a trend-following can make money in any direction but not during price trend transitions. 

Systematic trend followers often use stop-losses which create synthetic option pay-offs. There is a willingness to create small loses in exchange for holding onto potentially large winners. This behavior will create a positive skew pay-off.

Diverges are more likely when there is a crisis or some large market dislocations which means that the positive skew events are most likely to occur when traditional long-only risky assets are in decline. This makes a skewed lottery ticket that actually adds value. 

This "non-crisis alpha" performance has caught investors off-guard and has created confusion on whether this is a good time to invest. In fact, an equity crisis is not needed for managed futures to do well. There just needs to be an extended divergence or dislocation from existing price equilibrium for a broad set of markets.  

Managed futures are generating expected returns for the strategy, yet investor flows have not been as strong as might be expected. Some have the view that managed futures, a long/short strategy, is like a long-only investment that will mean revert and loss money for new investors. There is no reason to expect mean reversion with fund performance because any reversal in price can also be exploited. The risk is during transition periods not during periods of divergence. 

Tuesday, September 3, 2019

Yield curve inversion and stocks returns - Not too nice

Cam Harvey of Duke University was interviewed by his partners at Research Affiliates on the impact of the yield curve inversion on stock returns. The numbers are not attractive. Conducting a simple event study of the seven inversions since the 1960's, Harvey shows that excess returns are negative on average for three years. There have been yield curve inversions where stocks have moved higher but this is not something an investor should bet on. Only the 1978 inversion showed positive cumulative excess returns for all periods after the inversion. Two of the seven inversions ended with positive cumulative excess returns at the end of three years. 

The odds are stacked against equity investors. As researched by Harvey in his 1980's work, the real yield curve provides important information on future consumption. An inversion negatively impacts consumption growth, so the smart bet is to lower overall risk and focus on other asset classes or alternative investments. The numbers show that there is not an immediate link between inversion and stock declines but investors should realize that any delay in adjusting portfolio allocations will be costly. Unfortunately, the natural flight to bonds may also be a risky bet, so investors will have to think about a broader set of alternatives. 

Expected real returns are scarily low - What can an investor do when choices are limited?

What can we expect for real returns over the next few years? A simple examination of the GMO real return projections suggests that traditional assets will be give investors poor return choices. For pension funds, the chance of meeting their actuarial expected returns of around 7 percent seems nearly impossible. 

The only way to gain better returns is to get out of the slow growth developed worlds and hope that emerging markets will be a growth dynamo. It is possible that EM will be a help but this strategy will be risky. 

The simple combination of 60/40 stock/bonds in the US using large cap equities will generate a real return of approximately -2.9 percent. Even if the equity portion is diversified across international (DM and EM) stocks and the bond portion holds some international (EM) exposure (30/10), the real returns are not attractive at -.91 percent.  

A comparison can also be done with the expected real returns as generated by Research Affiliates. Their real expected return forecasts are for 10-years versus the 7-year used by GMO; however the story i very much the same. The real returns on both risky and safe assets are also very low, much lower than current expected return assumptions. RA has the expected real return at approximately 1 percent. Again, this is not going to reach any expected returns used by pension funds. 

Is there a solution to this problem? Certainly focusing on market beta will not provide good answers. Investors can look beyond classic equity and bond portfolios to include emerging markets, but perhaps a new paradigm is necessary. There are factor risks beyond market beta that can offer diversification and may not have the same level of overvalue. There is no easy solution but thinking beyond asset classes and dynamically looking at factor risks may offer some solutions.

Monday, September 2, 2019

What are the social and cultural influences on your investment decisions? It matters

Decisions can be classified by their level of predictability and level of social influence. Some decisions are made by the individual while others are made as part of a group or influenced by the choices and decisions of past groups. There is an evolutionary and cultural component on how we make decisions that is not often discussed. Many decisions are collaborative and have been influenced by how decisions were made in the past. 

This decision framework is presented in the short fascinating book The Importance of Small Decisions. They use a 2x2 matrix to how different decisions are made. Using a matrix between transparency and social involved to discuss how investment decisions are made can be informative and allow for some deeper insights on how to any improve decision-making. 

There is a continuum of decision transparency. The transparency is the link between action and result, the feedback between the decision and its consequences. If there is no decision transparency, there is no potential for learning and improvement. Skill cannot be acquired if there is no transparency between prediction, action, and result. 

Some environments are very transparent while others provide only a limited link between action and results. An environment that has very set rules is transparent. For example, a game like checkers is very transparent, but a game like poker is more opaque. Investment decisions are more opaque. In the simplest case, a decision to buy matched with a gain would be a transparent choice. Of course, there can be an error between any action and result based on a poor decision that happens to be correlated with the result.  

The other dimension of a decision is the social relationship. Some decisions are done alone while others are part of a group based on shared information, knowledge, or a sense of community. A committee decision is a social. Some individual decisions are mixed because the choices made by an individual will affect others. Hence, there is a social component.  

Each decision can be classified by the combination of social influence and transparency and can be placed in one of four quadrants. The second quadrant is where most decision choice work is done. Decisions in the third quadrant are guesswork. There is little chance for learning and corrective feedback. The first quadrant frames good inclusive decisions. The fourth quadrant can considered copying with little judgment. Action is taken because that is the way it has been done by peers in the past. 

Most of the time spent in finance and economics is in the second quadrant of rational choice. Decisions are transparent and just involve the individual, yet the majority of the decisions that we make are in the other quadrants. There is less work done on opaque decisions by individuals, which fall into the category of guesswork. Investment decisions often fall into this category because it is hard to match the decision with investment results, yet there is little direct discussion of guesswork. Still, when there is more guesswork, there will be greater desire to avoid making individual decisions. 

Often overlooked are the decisions that have a social component. Of course there has been significant work in game theory which discusses the interaction with others but there is also the more fundamental component of group decisions. Investment committees especially at larger firms are the norm and this makes these decisions a group process. There is a social component between the portfolio manager and the analyst which creates a group dynamic. Investors need to think about the cultural and social component of their decisions and determine whether they are helpful or a hindrance to their investment success.