Saturday, October 31, 2020

Back to basics - Always hold increasing returns to scale businesses

What should drive your equity choices for the rest of 2020? The classic factor style approach focuses on value, quality, size, momentum, or volatility which may often seem abstract from the underlying businesses. For example, value or quality is a factor which is sorted and not a strategy or a review of the business and the environment it works in. 

Simply put, successful long-term businesses that produce high returns should be those that have increasing returns to scale that can be exploited, serve as a tailwind, and can stop competitors. 

Most good firm will have at least constant returns to  scale. If a 10% increase in resource usage of labor and capital leads to a 10% increase in production, there is constant return to scale. If the output increases by more than 10%, there is increasing returns to scale. We can look at the classic Cobb-Douglas production function Q = K^a*L^b and work through the math to find conditions for increasing or decreasing returns to scale. Increasing returns to scale will lead to economies of scale or decreasing long-run average costs which support higher profits.

Increasing returns to scale is not the same as just getting large. Many firms think they should merge or acquire others to get scale by sharing supplies, management, or platform costs, but that is not the same as increasing return to scale nor will it create economies of scale with the underlying activities of the firm.

There are number of businesses that will have increasing returns to scale, those with high upfront costs or R&D, network effects, customer learning curves, branding for quality and location. Meaningful increasing returns to scale exist for businesses that create barriers to entry from their scale. For example, look for network scale. The attractiveness of a network product grows through the use by more customers, and more network users create value for all so that even moving to a cheaper product that does not have network is expensive. Firms that set an industry standard and gain a network are especially attractive and will have increasing returns to scale that will be hard to overcome.

However, these increasing return to scale business have issues of winner-take-all competition. There is a degree of gambling with finding these firms especially during their early life stage. Hence, there should be desire for those that have broken-out ahead of other competitors. Investors may be late to early gains, but they have the opportunity for sustained returns. 

The result of an increasing returns to scale business is high industry concentration which leads to demands for some regulatory control. We are seeing this now with the network tech companies. Government has to be large to monitor this concentration. There are political issues if the economy is dominated by large firms and large government for oversight. Small businesses may suffer because they cannot compete or be subject to a single service provider. There are spill-over effects beyond profit and loss but as an investment strategy it will always have merit.

Wednesday, October 28, 2020

Trend-following strategy as the new safe haven asset


Trend-following has often described as a "crisis alpha" strategy, but there has always been a simpler crisis alpha strategy over the last two decades, buying Treasury bonds. Given their strong negative correlation and positive total return during a market decline, Treasuries have all of the characteristics of a great crisis alpha strategy. That story may not hold going forward.

Treasuries have been viewed as the ultimate safe asset. Investors have the minimal credit risk from the US Treasury. Investors trade in the most liquidity market in the world, and if there was a crisis, fund flows would drive the price of the safe asset higher. Treasuries have been negatively correlated with equities; more negative than the correlation between equities and trend-following. The cost was low and in stable times you still received the coupon payments. What was not to like from using Treasuries as your hedge and crisis alpha?

Those halcyon days of Treasuries as a safe asset strategy are over. Liquidity can be a problem as seen in March. The Fed is now the largest holder of Treasury debt which changes the competitive nature of the market. Rates are low so you don't receive any yield especially inside 5-years to maturity, and the lower zero bound places a restriction on the potential for gain. You will not get a return kick to offset losses from an equity decline. Treasuries are not the safe asset as in years past. There is no total return offset to losses from risky assets.

So, what can an investor do to find an alternative safe asset? You can buy other forms of fixed income to offset the low yield, but safety becomes a question. Many other perceived safe assets have high risk characteristics so it is not clear they will provide safety when needed. 

Looking back in time, Treasuries have done a great job of providing portfolio safety and protection. Looking forward it is not clear that this will be repeated. On a forward looking basis, trend following may serve as a superior safe asset strategy. Instead of buying a zero return asset like bonds, investors should hold a strategy that will take positions based market price direction either up or down. Trend-following may still hold bond positions, but only if price direction warrants the exposure.

Buy the positive convexity strategy since the upside gain from holding bonds in a down market will not provide the upside needed to offset losses. Buy the diversified portfolio of trends because it offers a better mix of protection alternatives. It is a simple strategy. There are no guarantees it will do better in the short-run versus bonds, but in the current environment, it can still be the best alternative to a bond hedge.  

Tuesday, October 27, 2020

Trend-following - Deep learning can help but burden still on setting up the problem correctly

Deep learning techniques are a natural for quantitative investing and more specifically for trend-following, but just because the concept of deep learning is good for finding improved models does not mean it should be used. I am not anti-deep learning but realize that there are high barriers to entry to get it right as well as barriers to success based on costs of transacting. There is no simple answer with using these techniques because the value-added is all conditional on the set-up of the problem and the type of techniques used. 

A recent paper applying deep learning to trend-following shows that it can have a significant improvement in Sharpe ratio and other key statistics, yet there can be high variation in success based on problem formatting. See "Enhancing Time Series Momentum Strategies Using Deep Neural Networks"

No one should be surprised with this result. There have been hundreds of trend-followers through time that have all had their own variation on trend identification, position sizing, and portfolio construction. There is no single model solution. There is high variation in performance based on the techniques used and assumptions made. 

The researchers form a control strategy without any deep learning and then compare against four different approaches: 
Lasso regression (LINEAR)
Multi-layer perceptron (MLP)
Wavenet based on convolutional networks (CNN) 
Long short-term Memory (LSTM).

The techniques employ increasing complexity with LINEAR being a relatively simple linear model, MLP using a 2-layer neural net, wavenet  accounting for long history, and LSTM using sequence prediction based on memory structures.

These are some of the best techniques for use with time series data and are fitted against a number of objective functions such as Sharpe ratio, return, MSE, and binary choice. All returns are rescaled to a target volatility to ensure fair comparisons. The following two exhibits show the cumulative returns for the set of models. The box plots show the range of results across individual assets. There can be large differences between asset return performance across models.

Deep learning techniques can improve gross performance statistics, but it comes at a cost. More turnover translates to greater drag based on the transaction costs of constant adjustment. The impact of transaction costs is not trivial and can turn the great strategy into a marginal or losing model. The exhibits below show the added turnover and the impact on Sharpe ratio as transaction costs increase. 

Deep learning or any strategy with constant adjustments work effectively when transaction costs are not included or are very low. Increases in cost will have a much greater impact on deep learning models than simple approaches. Hence, transaction costs must be accounted for when building models. 

The challenges of trend-following research are the same across time. Improvements can be made through better and new techniques but costs matter and may lead the investor back to simple models.

Monday, October 26, 2020

Always a frustration when actual return is disconnected with volatility, but does it matter for managed futures

A frustration with any investment choices across asset management styles or risk factors is when the actual return to risk is negative. A basic premise for investing is that higher risk will be compensated with higher return. However, there is a big difference between "will be" versus "should be", a big difference between expected return versus actual return, and a big difference between risk/return trade-offs in the short-run versus long-run. 

When the return to risk is negative over a term of say three years, investors will generally avoid the style, or at least be more cautious concerning any decision for risky managers. The chart presented for managed futures is not unique to this return to risk problem. It just happens to be a focus for my current attention. I charted the return to risk for managed futures funds institutional classes listed on Morningstar for the last three years. Higher volatility did not guarantee any positive return. 

A contrarian may say this cannot last and choose the fund with the greatest return to risk versus longer-term averages, but a more conservative investor will just avoid the risky managers until there is some clear compensation for risk. 

Of course, for a defensive situational investment, return to risk may not be relevant but a mistaken measure of investment success. Managed futures (mainly trend-following) is expected to do better during market divergences, yet there is no reason why any rolling three year period return to risk has relevance for choosing a strategy that thrives during focused periods of market dislocation. Investors will pay a price for convexity; nevertheless, they should still be compensated for buying a riskier managed futures style especially if there is a period of market divergence.

Treasury liquidity and the demand for cash - There can be a run on the safe asset market

Investors generally have a strong view that Treasuries are a global safe asset. It is a true that they are the place to go if you need safety from risk. Demand will increase when the economy slows or there is a decline in risk asset expectations.  Some will say prices rise for Treasuries during a crisis because there is a shortage of the safe asset. Having a safe asset is important, but Treasuries are not cash.

At some level of uncertainty, not just risk, investors will want cash, the ultimate safe asset. At high uncertainty there is no demand for an asset with duration risk, no desire to hold an asset that will have a charge for liquidity through the bid-ask spread, for some settlement uncertainty, or has pricing uncertainty based on short-term flows. As uncertainty rises, Treasuries become just another risky asset. Perhaps less risky than other assets but not a safe asset. While this is not the conclusion of the Fed, it is a reality when reviewing the deep dive on Treasury liquidity by the NY Fed as presented by Lorie Logan in her speech "Treasury Market Liquidity and Early Lessons from the Pandemic Shock", Remarks at Brookings-Chicago Booth Task Force on Financial Stability (TFFS) meeting, panel on market liquidity (delivered via videoconference).  

When the pandemic policies hit in March and there was maximum uncertainty on the course of the US and global economies, there was not a flight to Treasuries as the safe asset. There was an unwinding of any risk and a move to cash, a run on the safe asset. Hedge funds sold levered positions, foreign central banks raised cash, and mutual funds sold risky assets to prepare for outflows. When there is an investor run on risk and a desire for cash, the safe asset will be subject to distortions and become unsafe. None of this should have been surprising when the herd has a massive change in expectations. 

The Fed had only one course of action; become the buyer of last resort. However, Fed purchases of Treasuries have continued, and the crisis manager is now an integral part of the safe asset market and not just a lender of last resort. 

The real question is what would the Treasury market look like today if the Fed was a less significant player? The technocrats are now running the market, so hope that they have the vision to deal with the known unknowns and unknown unknowns that we can face with the safe asset. These choices do not belong to the private market.

Thursday, October 22, 2020

Forget about new agriculture futures contracts; Focus on agriculture competition if you want to trade commodities

A new CME pork cutout futures contract is coming in early November. Why should a commodity investor care?  Most new futures contracts fail. We have not seen a new successful agriculture futures contract for some time. Most investors will not be an active user until volume and open interest rise to some minimum level to sustain speculative trading. 

What is important for the success of a futures contract is the underlying structure in agriculture markets and price discovery cane be facilitated through futures. An industry that is too concentrated cannot sustain active futures contracts. When there is a power imbalance between buyers and sellers, the dominant player will drive contract decisions and pricing and will generally not support open discovery of prices. A poor functioning cash market because of extremes in market power will spill over to a poorly functioning futures market. 

A less competitive market means that traders are not able to see or be part of price discovery and futures prices are not closely linked to fundamentals. Even a trend-follower who focuses just on prices does not benefit from markets dominated by larger players or a market where price discovery is not present. 

Agricultural markets are becoming increasingly dominated by large product buyers which impact speculation, hedging, and overall prices. Grain markets are also dominated by a few players in farm production inputs like seeds. One-sided relationships will threaten competition. Yet, there are limited voices for competition given smaller players do not have the ability to influence exchanges or regulators. 

However, the story of competition in the hog market is more complex than a David and Goliath story between large processors and small farmer/producers. Processing has become more concentrated, but hog producers have also become big operations that need stable pricing and contracts. Direct placement over auction is more likely when both parties are large and have a desire for consistent product flow and stable cash flow. The changes in the hog industry is a story of increasing returns to scale with vertical integration to ensure scale economies are exploited. Processors buy or link with producers to create a seamless supply chain. In this world, the needs for a lean hog futures market change. 

There is a desire to switch futures business for pork cutouts over lean hogs because the business has changed. The key firm risk has moved to a different point of the supply chain. The combination of concentration, the evolution or destruction of the cash market, and the value-added at the wholesale market over the risk to farmers all contribute to a need for adaptation in futures contract, but it begs the question of what is happening to agriculture markets.

Concentration has slowly increased over a 40 year period and now is closing in on 70%. The concentration has coincided with the export business for protein which takes greater capital and logistical expertise to be profitable. However, these numbers do not tell the real story because large hog producers are also dominating the business and eliminating the small producer. Direct contracting is dominating the market over a cash spot market. An integrated hog industry changes the need for generalized futures contracts.

There is a dynamic tension between integrated market structure, competition, and futures trading. Futures trading sits in the center of a competitive industry structure. Change the rules and players, and the active contract of today can be a dead market tomorrow. Agricultural futures have now survived for centuries, but its importance in farm marketing and distribution is not guaranteed as the move to industrialized farming moves forward. 

Wednesday, October 21, 2020

Has 2020 been a good or bad year for trend-followers?


Has this been a good year for trend-following CTA's? The classic approach is to just look at the time series for an index of trend-followers. You can use the SocGen, BarclayHedge, HFR, Eurekahedge, Credit Suisse, or Nilsson Hedge to name just a few. Some are equal-weighted, while others are AUM-weighted or restrict an index to large firms. Not all firms report to all performance services so there will be some sampling error differences. 

I looked at some slightly different metrics for comparison for this year because there has been a significant amount of dispersion between managers this year. Dispersion can lead to an index being centered around zero yet there being some clear performance winners. I am not conducting an exhaustive analysis but focusing on some interesting numbers that provide a different perspective. I focused on the Nilsson Hedge database and sorted on the largest reporting trend-following CTA's. The sample includes 52 names of funds greater than $200 million. There may be multiple funds from the same manager. 

The average year to date return for 2020 is -50 bps while returns for the same period last year was 10.17 percent. The sharp swings during the early pandemic took a toll on managers. Less than 25 percent of the managers showed gains of more than 5 percent while over 70 percent showed gains above 5 percent last year. Funds over a billion dollars in AUM showed slightly lower performance (-73 bps) and there was only a one out of ten chance of exceeding 5 percent in year to date return.

Compared to historical performance, no fund has done well even with a 30 percent difference between the high and low fund. Comparing long-term average annual return plus a standard deviation for each fund shows no fund above that threshold. 45 percent of the funds were at least one standard below the average annual return. 

There is still one quarter left in the year, so performance can change especially given the uncertainty in markets, but this has not been a "crisis alpha" year. 


Tuesday, October 20, 2020

What to expect as the risk-return trade-off for trend-followers


What is the expected return to risk trade-off for trend-following CTA's? Using a cross-section of trend-followers above $200 million in AUM that report to the Nilsson Hedge database, a regression analysis finds the return to risk trade-off is stable at approximately .6. Increasing from 10 to 20 percent volatility will raise expected return from 6 to 12 percent. Any investor who would like returns above 10% will have to lever a fund or find a manager with volatility between 15 and 20 percent. 

Of  course, there is variation away from the regression line, but a review of the 52 trend-followers in the sample tested show Sharpe ratios that range between just over 1 to a slight negative value. 

Using this information as a basis for comparison, a trend-following program that can consistently have a Sharpe ratio above one is exceptional. Clearly the rolling Sharpe ratio can move higher and lower than the range shown, but as the sample size increases, the average Sharpe is likely to fall within the range calculated. 

To get a higher Sharpe ratio a manager will have to blend other styles with trend-following to smooth the return and lower volatility. This is not an indictment of trend-following but the core strategy will have a long-term Sharpe ratio as a base. Improving on the base requires improvement in signal extraction, portfolio construction or risk management. Next to model improvement, adding other alternative risk premia is the most likely method for Sharpe improvement. All of these changes will have an impact on trend-following convexity which is the core reason for holding this strategy. These are the trade-offs that have to be considered by investors.

Radical Uncertainty - we are living it, so learn to deal with it

The best book for the year and should be on anyone's pandemic reading list is Radical Uncertainty by John Kay and Mervyn King. It tackles a timely problem that keeps any academic or practitioner up at night, uncertainty. Managing uncertainty, not risk, has been the leading challenge across my career. How do you make effective decisions when faced with uncertainty, not measurable risk? Uncertainty was one of the key issues for Keynes, Frank Knight, and many of the leading statistical minds of the last century.    

This book is a wide ranging discussion, yet it is not a cookbook that will give you answers for how to solve problems of uncertainty. It presents the many facets that are associated with measuring or forming expectations on events that are not easily measurable and cannot be counted. The countable or frequency of events allows for a measure of risk, but reality is filled with ambiguity and vagueness given the uniqueness of future events. Nassim Taleb has focused on the extreme unknown unknown, but there are also more situations closer to home which can be imagined but difficult to handicap. We can obtain measures of price movement like standard deviations, but these does not help us with thinking through events that are not countable and require subjective measures in order to form actionable decisions.  

Radical uncertainty, measuring hard to assess future events, requires a deeper understanding of scenarios and choices that cannot be merely expressed by numbers. We often have to imagine and form narratives and stories that are logically consistent to handicap and prepare for the future. The successful investors during this pandemic were able to imagine  radical futures not seen by others. It is a skill that is necessary and can be learned.

Monday, October 19, 2020

Diversification the driver for smart beta returns

Most of the outperformance of smart beta can be explained by diversification returns embedded in rebalancing. The value-added is just good old diversification and rebalancing. Keep it simple. This conclusion is the interesting result from the paper "Is smart beta still smart under the lens of the diversification return?" by Lin and Sanger that has finally been published.  The work should make any investor think twice about paying a fee premium for some of these smart beta products.

The foundation behind this work is from older research developed in the early 1990's by Booth and Fama and Fernholtz and Shay in the 1980's on diversification returns. The average compounded return is the weighted average geometric returns from the stocks in the portfolio and the diversification return which is half the different between the weighted sum of individual stock variances and the portfolios variance. This return is a portfolio version of geometric returns which is the arithmetic return minus the variance drag. This return calculation concept is learned early in the study of investments. 

Given this simple concept, the diversification return effect can be measured. An improvement in diversification return can occur if both portfolio and individual asset variances are controlled or minimized. The authors look at different weighting schemes applied at the industry and at the stock level inside an industry grouping. These can be used to make comparisons with the classic value-weighted index and equal weighted within and across industries. The set of alternative weighting schemes are included in the table below.

The figure below shows the results from the author's analysis. The cap-weighted index underperforms the other weighting schemes based on geometric returns calculated over the long period tested. Using alternative weighing scheme reduces risk of large losses and improves the Sharpe ratio. The driver of returns is the diversification rules.

 We have recently posted "Thinking about diversification in an uncertain world" which focused on Charlie Munger's comment that diversification works for the "know-nothing" investor and that the goal of investing is to find investments where it is safe not to diversify. This smart beta work still reinforces that argument. It makes perfect sense to push the value of diversification as far as possible. Get as much free lunch from diversification first and then use skill if available to take specific risks. 

Strong cap-weighted index performance this year does not change the core results for this study. The big tech names have dominated the economy and asset recovery, but in a more normal world diversification will still win over the long-run. 

Thursday, October 15, 2020

Understand r-star if you want to understand the Fed

The Fed has policy interest in r-star, the rate of interest that will occur when the economy is at full employment and the targeted inflation rate; however, it has solicited mixed reviews in Fed speeches given it is an unobservable number and subject to distortions when rates are low. 

It makes perfect sense in theory to serve as a guide if the estimates are good, but that is a big if. If rates are higher than the appropriate r-star or neutral rate, monetary policy is not accommodative, and the Fed should use policy to lower rates. Similarly, if rates are below r-star, the Fed should tighten policy to offset the accommodative environment. 

Currently, the two r-star models posted on the NY Fed website suggest that the equilibrium real short-term rate is below 50 bps. By this measure, Fed policy is accommodative but therein lies the problem with r-star as a policy tool. If r-star is not measured correctly, there is the potential for wrong-footed policies. The Fed may think it is accommodative, but in reality, it could be following a tight policy. The r-star models can tell us something about the neutral rate equilibrium but it should not tell anything about policy moves.   

A new paper, "Estimates of r* Consistent with a Supply-Side Structure and a  Monetary Policy Rule for the U.S. Economy"suggests that the problem of estimation error with r-star is real and should be a concern for policymakers interested in using it to guide monetary action. The paper makes enhancement to r-star through adjusting for the zero lower bound on rates and accounting for the relationship between rates and the IS curve. The lower bound problem can be addressed through forming shadow rates which can go negative. Changing the model generates significant differences in r-star estimates

The results show that r-star is currently negative and heading lower. This means that current nominal rates minus inflation may actually be less accommodative than thought. Inflation has to pushed higher to generate a real rate that will get us to full employment. The devil is in the details, but the changes in model assumptions and structuring will get different results. It suggests that r-star guide may be suggestive but not a helpful guide for policy.

Still, following r-star can help asset manager as a measure where rates should be over the longer-run. All r-star models show that the equilibrium real rate should be lower today than over the last few decades. There is nothing that will suggest rates should be headed higher or that the Fed will take action that will offset current accommodations. The Fed has given forward guidance, but r-star provides empirical evidence that rates will stay low. As a blunt instrument, r-star can still be helpful.

Wednesday, October 14, 2020

Political Risk Assessment - Big exposures are often under evaluated

Macro investing will focus on global monetary policy and growth and make assessments on relative asset returns, yet there does not always seem to be a structured approach to political risks. A structured approach looks at different types of political risk and then tries to understand the core questions for these risks, analyze the likelihood of these risks, form an action plan of taking exposure or mitigation, and then responding with trades. 

These issues are addressed in the book Political Risk: How Businesses and Organizations Can Anticipate Global Insecurity by Condoleezza Rice and Amy Zegart. While this is a book focused on corporate political risk assessment, it is very relevant for any fund manager. 

The book focuses on a set of political risks and when I review the list, it becomes immediately obvious that it is applicable to any market assessment.  Some of these risks are firm specific and not asset class focused, but it still has clear relevance for macro investing.

The problem with any political risk assessment is that many of these risks have low likelihood of occurring, so it is not easy to price and take advantage of opportunities. Most political risks have the chance of being large tail events given they are hard to price; consequently, it is all the more important to analyze and assess. 

Rice and Zegart focus on a four step process: understand, analyze, mitigate and respond. Understanding political risk starts with assessing a firm's political risk appetite. For example, an investment in Mexican equities or bond is a relative return analysis based on political risk assessment. If you cannot or do not have the skill to take on political risk, then it should be avoided regardless of the potential return. The analysis starts with the information that is available. Is there good information on the risks that can be used for rigorous analysis. Risk mitigation is a function of having a warning system that can alert you of higher risks. The warning system will allow risks to be mitigated and damage limited. Finally, effective risk management will assess responses to threats. Good response determines whether risks were avoided, and opportunities capitalized.

If you are a macro investor, a structured assessment of political risks is essential for good portfolio construction. There is no positive gain from diversification if there are just hidden political risks unassessed. 

Tuesday, October 13, 2020

LEGO can tell investment people something about risk management


Investment management has not cornered the market on risk management strategy and techniques. Investors can learn from how firms approach corporate strategic risk management. One firm that seems to be ahead of many is LEGO. It has been successful navigating geopolitical risk, competition, and changing tastes, to become the largest toy firm in the world and the envy of all for their good values. 

The LEGO approach to risk management can be summarized in an informative article, "Strategic Risk Management at the LEGO Group" Strategic Finance, February 2012What I find especially appealing are two LEGO ideas associated with their risk management process, Active Risk and Opportunity Planning (AROP) and the Park, Adapt, Prepare, and Act (PAPA) Model.

The AROP process subjects every project to a comprehensive review and analysis of not just downside risks but also upside opportunity. Good risk management does not dwell only on the bad things that could happen from undertaking a project but also looks at potential gains. Risk management without assessing upside is just an exercise in risk mitigation, and the easiest way to employ risk mitigation is to not take any risks; reject everything. AROP analyzes both tails to ensure appropriate risk assessment.

The PAPA model is a classic 2x2 matrix which looks at the firm's strategic response based on likelihood and the speed of change of different scenarios potentially being faced. This mapping of project or strategies into scenarios is a good way of prioritizing projects, opportunities, and risks. 

Scenarios that have low likelihood and a slow speed of change can be parked, although not forgotten, until the future, while scenario events that have high likelihood and fast speed of change will likely require immediate action. High likelihood events that are slow to evolve will involve adaption while fast events or scenarios with low likelihood require current preparation.

Investing has not cornered the market on good strategy and risk management principles. Investor can discuss different world views, but those views have to be placed in some context and looking at likelihood versus speed of change is a strong actionable framework. The AROP work can be mapped into these  strategic scenario assessments of the economic and market environment, so there is a holistic approach to test and compare investment ideas. Thoughtful analysis can lead to repeatable success if it is implemented in a structured framework. 

Monday, October 12, 2020

Thinking about diversification in an uncertain world


"The "know-nothing" investor should practice diversification, but it is crazy if you are an expert. The goal of investment is to find situations where it is safe not to diversify." - Charlie Munger

What is the best strategy if you are uncertain and not sure what will happen with the US election and the global economy? Diversification. What is the worse strategy if you have a view on the election outcome or the global economy? Diversification. 

There can be too much diversification, and the level of diversification should be viewed as dynamic. Diversification should be assessed on a regular basis based on the investor's knowledge and market environment. Similarly, the type of diversification employed should be regularly assessed. For example, holding a cap-weighted benchmark may not provide the diversification desired. 


Diversification - You need "intelligent supervision"

A simple taxonomy of diversification - All diversifiers are not alike

A primary goal of effective investing and portfolio management is to find the investment where you may have an edge. This is not a perceived edge but a real edge or view that you are willing to commit capital for as a risk. This does not mean there is certainty with your bets. Your edge is a subjective expectation for gain relative to risk. 

The problem with using any subjective expectation is that you don't have good feedback on the quality of your thinking except through measuring the results in your portfolio. A novice, by definition, should take smaller bets and have greater diversification because the measure of your subjective success is limited. As the quality of your subjective expectations increases, the degree of diversification should decrease. If there is no measurable skill, high diversification should be maintained. 

Diversification is a management tool that should change with your skill at assessing the current environment. You can be a skillful investor, but perhaps not under current conditions. Hence, today may call for more diversification.

Assess macro-prudential policies and tail risk


The focus on tail risks cannot be separated from Fed and government macro-prudential policies, the global and domestic financial safety net. Measuring the safety net is a critical input in any tail risk analysis. What will be the policy response of the central bank and government? What will be the speed of the response? What will be the size of the response? Tell me the type and form of the safety net and I should tell you the amount of tail risk I will face. 

The March liquidity crisis is a perfect case study. The tail risk was significant but upended and reversed through government bond purchase and the announcement of other market support mechanism. For the investors who got overly defensive in late March, it was return disaster. For those who rebalanced at the end of month or held their positions, it was a time huge success.

If the government will respond swiftly to arrest any strong market decline, the amount of private insurance or hedging can be reduced. If the government will supply and bear the cost of the safety net, how much extra or reinsurance should I buy? Accounting for these policies is making the Joseph Stiglitz phrase of "socialize losses and privatize gains"  operational as part of your tail analysis. 

There has been research on how macro-prudential policies should be counter-cyclical, yet in reality, we are seeing "pre-emptive" monetary policy to stop market declines early in an effort to reduce any negative wealth effect on the real economy. Hence, we are seeing rates stay low longer in an effort to ensure stable wealth to further support any recovery.

An argument that has been going around for decades is that the polices of the Fed, starting with the "Greenspan put" and currently with the "Powell put", have shortened and reduced the amplitude of any financial crisis and thus reduced the benefit from holding defensive strategies like trend-following managers. This may be true, but it can be uncertain whether future policies will remain static.

The question for tail risk management is the level of decline at which the Fed will support the market. Of course, the fiscal side and policy can also be used to support financial prices. In other countries, we have seen intervention in market rules such as banning short sales. A simple analysis or simulation without including policy choices will generate a false narrative. 

However, policy risk and uncertainty have to be modeled. What happens if the government changes its reaction function? What if the Fed does not move to support markets quickly, or even worse what if the policy tools used to support the markets are less effective? While there really is no constraint on the Fed balance sheet, the combination of a zero bound and muddled forward guidance can change market expectations quickly. 

Whether you like it or not, tail risk analysis requires a policy reaction, safety net, function both for the upside and the downside tail. 


Friday, October 9, 2020

CNN Fear and Greed index consistent with other risk aversion and stress indices

Focusing on stress and market sentiment indices before the election has been useful for providing evidence that is contrary to many talking heads. The reading of the CNN Fear and Greed Index which is a variation on the stress indices we follow shows levels that at 50 right in the middle of the range. There is a balance between greed and fear measures.

The CNN index consists of seven indicators that are more associated with the stock market. The seven include: 
  • the junk-investment grade spread, which is tight and indicating greed;
  • market momentum as measured by deviations from the 125-day moving average, the market is high above the 125-day average which suggests greed;
  • the put/call ratio, which is showing low values and also suggesting greed; 
  • the VIX volatility index versus the 50-day moving average, which is showing a neutral indication; 
  • the safe haven measure of 20-day stock/bond relative return, which is low and showing fear.
  • the stock price breath as measured by the McClellan volume summation index, which is indicating fear given its low level;
  • the stock market strength as measured by the 52-week net new high, which is at the lower end of the recent range and also suggests fear.  
There is a balance between greed and fear versus what existed last month. The market does not suggest negative sentiment or significant stress. We have looked this week at the CSFB Fear index and the OFR Financial Stress Index and have gotten similar readings. See: CSFB Fear Barometer - Why is fear stable? and No stress in the markets, but it has a way of sneaking up on investors.

Thursday, October 8, 2020

Does your investment committee have a Devil's Advocate?

Does your investment committee have a Devil's Advocate? A Devil's Advocate is someone who takes the opposite or alternative argument to dispute the conventional or majority opinion.  Not many want this job, who wants to be the devil, but it can be critical for the firm if it wants the best possible analysis. 

The phrase has a long history first associated with the Catholic Church in the 16th century. The Church would review the information for sainthood of an individual, but they decided it was important to have someone take the other side of the argument and dispute the case for canonization. This theologian or officially appointed skeptic was a "Promotor Fidei" or promotor of the faith; however, given the task of advocating against canonization, the person became known as the advocatus diaboli—the devil’s advocate. In 1983, the Catholic church diminished the role of this advocate.

The Devil's Advocate does not have to believe in his position. The focus is on getting an alternative view to the current position being advocated and questioning the logic and supporting data for a position. If someone argues that equities should move higher over the next six months, the Devil's Advocate will marshal facts and theory to argue the opposite opinion. Even a quant firm should have a Devil's Advocate who will argue against a new model being presented or dispute the quality of the back-testing supporting a model. 

This is not an easy position to hold. Hence, this role should be rotated over time. You are not going to win friends or influence people if you are taking the unconventional view. In some firms, it may seem as though this type of advocate is not necessary given a culture of debate. However, having an appointed advocate against convention or the majority view ensures that a lively and healthy debate is possible. Good thinking only comes from strong debate. 

"If everyone is thinking alike, then someone is not thinking." General George S Patton. 

Wednesday, October 7, 2020

No stress in the markets, but it has a way of sneaking up on investors

Following financial stress indicators is a good way of cutting through the verbal rhetoric and watching what markets are really doing. The current reading shows a low stress environment, not as low as a year ago before the pandemic, but significantly below spring levels. Stress measures have been stable for the last two months.

There are financial stress indicators from brokerage firms and banks. Additionally, they are available from Fed banks and other government institutions. The Office of Financial Research (OFR) calculates a daily financial stress index (FSI) using 33 financial variables divided into five categories: credit, equity valuation, funding, safe assets, and volatility. The variables are all publicly available and the categories values are also published. This FSI also shows stress by region. The index is zero when the average is zero.

The credit category focuses on corporate spread levels which have stabilized since spring. The equity valuation is centered on price to book ratios. The funding section has seven measures of short-term dislocations like 3-month LIBOR-OIS spreads. The safe asset category focuses on demand for assets like Treasury bonds, gold, the dollar, and yen exchange rate. The volatility category includes volatility measures in stock, bonds, currencies, and commodities. 

While the current index levels are showing below average stress, a close look at historical data is that stress indicators shift quickly with limited advanced warning. A switch from negative to positive values should be a concern. A decline in stress is usually associated with Fed action to directly offer stabilization. For the near-term, investors should focus on near-term stress from election concerns and not poll numbers or talking heads. 


Tuesday, October 6, 2020

CSFB Fear Barometer - Why is fear stable?


There have been significant discussions about fear and uncertainty in the current marketplace, but it seems that to get a true idea of fear is to look at what is going on with market prices. Talk is cheap. Following action is more valuable. 

An innovative approach to measuring the action from fear is through the CSFB Fear Index which looks at the pricing of a zero cost collar on the SPX index. The index measures the premium of a 10% out of the money call and then finds percentage out of the money strike for a put on the SPX index that would make the combination a zero cost collar. 

If there is more fear, then an investor would only be able to buy a put further out of the money with the call premium. The fear index value would go higher. If the index falls, this fear barometer is telling the market that you can buy protection closer to at the money with the premium from the call writing. Since the index is pricing a zero cost collar, this index will account for any skew in option prices.

The current index levels are surprisingly stable over the last month in spite of the market sell-off. Even with election uncertainty, the fear index has been stable. The index is much lower than summer levels. The maximum fear was in February before the March liquidity crisis. Fear was actually at its lowest levels in March.

The long-term index shows that fear has declined from highs in 2016. The index is still elevated versus pre-GFC, yet the trend has been lower. It is interesting that fear has been growing or high for most of the period of equity gains. Fear and price moves have an interesting link that does require further study. Given the complexity of the relationship between equity index returns and fear, this index has not been given much attention, yet it provides a unique assessment of market opinion. 

Currently, the option collar is saying we have less need for worry. It is unclear why options have this relatively higher optimism.

Monday, October 5, 2020

Thinking through clustering for a different but clearer perspective

There are a couple of visuals that are always used in the money manager's toolbox. One tool for quick comparison is the scatter plot of return and risk across different asset classes. Some analysts will get more sophisticated with this visual by looking at how asset classes have moved through time in risk and return space. Some with better visual dexterity will look at three dimensions and include correlation. 

The visual information of risk and correlation is computed and collected in the covariance matrix which is a core component of any optimization. Unfortunately, the covariance matrix can be difficult to work with as more correlated assets are added to the matrix or if there is instability in covariance through time. The impact of covariance sensitivity is less intuitive on risk measurement and asset weight selection but is a critical part of asset allocation. Any optimization is sensitive to multi-collinearity and the difficulty of inversion. If there are more assets that have similar covariance, small changes in statistical characteristics will lead to significant changes in optimized weights; the optimized asset allocation is unstable. 

A tool that can be useful and that is easily visual for money managers is cluster analysis through the use of principal component factorization. Principal component analysis is a data dimensionality reduction tool. It looks for similarity across data or common feature extraction. By eliminating common features, we can find uniqueness. By grouping common features, we can find clusters which can be graphically displayed. 

When assets cluster around common factors or have similar covariance characteristics, asset allocation becomes more difficult. One, there is less diversification benefit. A cluster of assets around principal components will not offer any benefit to investors. Portfolio risk is not diminished. Two, the covariance matrix becomes less stable and optimization becomes harder. The asset allocation will become sensitive to small changes in the price behavior of any asset in the portfolio.

There are quantitative methods for finding clusters and adjusting the covariance matrix, but a first past is to focus on the intuition of cluster analysis. No different than any good data work, plotting the information is a critical first step for analysis. 

As shown in the graph above, there are some well-defined asset clusters and some assets that are unique. Assets in the clusters are going to add little value to the portfolio. Optimization across those clustered assets will shown unstable weights. Assets outside clusters will have strong diversification benefit. Assets outside clusters are more important to a portfolio and should be a place of focused investment effort. Look for commonality of factors and find uniqueness. This process of combination for commonality and uniqueness will always be rewarded.