Times series (Fama-French) or firm characteristics for measuring factor risks - Look at the firm characteristics

Investors have become more sensitive to measuring factor exposures in their portfolios; however, this measurement is not easy. There are two major approaches to measuring factor exposures, time series analysis and analysis based on firm characteristics (cross-sectional). Both have benefits and drawbacks; however, focusing on firm characteristics can provide more timely and detailed insights. 

The times series approach has been extensively researched and used by investors. As developed by Fama and French, the times series approach regresses portfolio returns onto risk factor returns to measure the statistical sensitivity of a portfolio to well-defined risk factors like size, value, quality and momentum. These well-defined risk factors can be easily found in the Ken French website.

The alternative approach is to look deeply into the firm characteristics or firm descriptors which proxy for style factors that can be further grouped into more traditional factors like value, size, and volatility. This cross sectional approach has been extensively developed into well-defined frameworks, (see, for example, MSCI  "Measuring Factor Exposures"). In the characteristic approach, the focus is on current holding exposures that translate into factor risks. 

A portfolio can be decomposed into the descriptors for a factor which provide details on the composition of the portfolio. This fundamental composition can be compared with the times series factor exposures and the factor exposures associated with these descriptors. A close examination shows that the risk factors from cross-sectional analysis map better than times series with the actual descriptors of the portfolio. A cross-sectional style factor for value will do better than a times series measure at representing actual real time risks.    

The time series approach is easy to implement; however, there are some problems that are hard to solve. If there are large and sudden shifts in the portfolio exposures, the time series approach will not pick them up. Time series risk measures will be sensitive to set-up specifications. For example, result will change with using weekly or monthly data and whether measurement is done over one year or three years. There is the potential for spurious correlations from these regressions. 

Characteristics, on the other hand, need detailed information on the portfolio and a large database by which to measure risks. If an investor does not have fully transparency of portfolio holdings, it will be hard to obtain a good measure of risks. For internal risk management, the characteristics approach is far superior to the alternative. 

The characteristic approach is foundational for forming an alternative risk premia index. For example, a quality portfolio can be formed through forming a weighting scheme of quality firm characteristics. Differences in quality portfolios can be measured through different characteristic weights. 

The Fama French time series changed thinking about on a number of levels, but as our ability to gather and analyze data has increased, alternative approaches have been commoditized and made easier to employ.

Yield curve inversions do not tell us anything stock market returns

Yield curve inversion has been the tool of choice for predicting a recession in 2019, but for an investor, the real question is what will happen to risky markets. At the extreme, who cares whether you can predict a recession, an investor wants to know what will happen to their equity portfolio. Fama and French have published a simple working paper on the impact of inversion on equity risk premium called, "Inverted Yield Curves and Expected Stock Returns".

Fama and French conclude that inverted curves do not predict low stock returns using a global database includes decades of inversions. They test the impact of inversion through a simple process of switching from a passive equity portfolio to a blended equity and cash portfolio based on prior time spent with negative term spreads. If negative term spreads (inversion) tells us nothing about the equity premium, then switching to a blended (active) portfolio will underperform the passive portfolio. 

The message is counterintuitive but clear. It does not matter what an inversion may be telling investors about recession, you don't want to bet against the equity premium. I would like to see more details concerning specific cases, but the unambiguous results using a simple test are compelling and will require deeper study to refute.  

Fiscal policy in current account surplus countries will spillover to deficit countries

The big trade and global structure trade for the next year has nothing to do with tariffs, but with changes in fiscal policy to address the savings imbalance problem. Tariff wars may push countries to change domestic policies to solve the savings problem but the line of causality is indirect.

The global savings imbalance still exists, and some analysts have been talking about this for well over a decade. This was formerly called by Ben Bernanke the "savings glut" problem. The North-South oil imbalance has been solved but the East-West imbalance still exists. The excess savings of Northern Europe and East Asia funds the current account deficit of North America. This will not be solved through raising tariffs. Tariffs will change the composition of trade across partners but will not eliminate the trade imbalance.

The savings imbalance (current account surplus) in Germany can be reduced by increased government spending through more aggressive fiscal policy. The China savings imbalance can be reduced by further switching from an investment led economy to a consumer driven economy. The savings imbalance can also be solved through tighter US fiscal policy, but this is an unlikely political solution. These fiscal and domestic changes are not immediately likely to occur, but if the signs start to exist, financial markets will react.

The impact of the changes in savings imbalances will spill-over to the US rates market. A reduction in global savings will place upward rate pressure in the US. This pressure could be contained through monetary policy which will impact the dollar. Nonetheless, these large structural changes will be the pressure points for where there will be the greatest opportunities.   

DM and EM equity market differences and commonality

The Norwegian Ministry of a Finance commissioned a study of global equity markets from MSCI, “Selected Geographical Issues in the Global Listed Equity Market”. This exhaustive research study on global equity markets could be better called, “The differences between developed and emerging equity markets”. This is valuable reading for understanding the return differences and similarities between developed and emerging markets.

The world is becoming more global and emerging markets are a more important part of equity markets, yet their representation is still lower than market capitalization and GDP would suggest. Investors need to have an EM opinion and have to think about global economic impacts on equity valuations. Trade wars have had a significant impact on equities because more global corporate sales comes from outside a stock's home country.

Returns are higher in EM markets but so are volatilities. Emerging markets are more diverse than developed markets. EM equities will be correlated with common factors like a financial crisis, but if an investor wants more diversification benefit, it will come from holding EM markets and not diversifying across DM markets.

The world equity markets can be divided into regions include North America, EMEA, the Pacific, and EM. These regions will show common return performance dominated by their largest country equity market. Still, market fundamentals, whether price to book, price to earnings, dividend yield, or return on equity, will generally move together. 

The EM sector is getting larger for two reasons, economic growth in EM countries have been increasing relative to DM so EM represent a greater share of global GDP and the fact that more countries and names are being added to the EM index. EM markets are more sensitive to economic growth especially with respect to the downside. There is a similar pattern of sensitivity of DM and EM markets to inflation.

EM stocks are also more sensitive to currency changes. The increase in the dollar has been an important drag on EM overall returns. The currency impact on DM equities is far less dramatic. It is necessary to have a currency view when investing with EM equities.

The equity risk premium is actually well-defined across countries with liquidity, governance, and size having additional impact on explaining this risk premia.

To be an effective global equity investor, someone needs to understand the similarities and differences across geographical regions, as well as between DM and EM markets. There is much commonality which leads to high correlations, but there are also enough differences to have it worth holding a diversified global equity basket. 

Market Efficiency and Hard Thinking -

Warren Buffett once thanked the disciples of the efficient market hypothesis, referring to them as “opponents who have been taught that thinking is a waste of energy”.  - Shareholder letter 1985

Thinking is hard; that is why we do so little of it. Thinking requires looking for links between the past and predictions of the future. Sometimes these links are present. Sometimes the links change in unexpected ways. Information gathering, which is the first step in finding a link, is hard especially if you are looking for something new in the data or for new information. Most times the links between information are poor and filled with noise, yet there has to be some investors who work at making markets efficient. Most will not be successful, but the process is critical. 

The paradox of market efficiency - 

• If everyone thinks the markets are efficient or there are no rewards from research, then no one will gather and use information and the markets will become inefficient.

Some corollaries -

• If everyone uses the same bit of information and model, the speed of adjustment or reaction to any new information release will increase and the value to the marginal investor is zero. 
• Market efficiency is about both being right and understanding behavior quickly.
• Passive investing cannot make markets efficient. 
• Passive investors are information agnostic.
• If you do not have an information edge, it is better to be information agnostic
• Most perceived information edges do not exist.
• Passive investors are asking others to make markets efficient.
• Active investing does not ensure market efficiency. 
• Active investing will not always be rewarded with greater returns.
• The mix of economic agents trading in a market affects market efficiency.
• Smart investors are not guaranteed success.
• Being smart does not mean an investors can decipher market behavior or profit from it. 
• Less smart investors will not always be eliminated from the market.
• Smart money does not drive out not so smart money. 
• There are limits to arbitrage 
• Making markets efficient from arbitrage opportunities requires capital and this capital may be finite.
• Central banks (non-profit maximizers) will not make markets efficient and lead to inefficiencies.
• Hedgers may not be profit maximizers.
• The size of players who are not profit maximizers varies by markets so efficiency differs across markets. 
• Beliefs can be rational but can be wrong and lead to inefficiencies.
• The can be many rational beliefs but only one market reality.
• Just when investors figure the market out, there will be a new problem.
• There is no market end game.
• If there are fewer analysts following a market, there will be greater opportunities.
• More analysts and capital following a market makes the speed of adjustment faster.
• Behavioral biases can be contained but not eliminated
• Bad behavior leads to inefficiencies and more uncertainty will increase bad behavior. 
• Efficiency is a dynamic concept. The efficient market of today may be inefficient tomorrow.
• Risk premia will still exist in efficient markets 
• Markets are often micro efficient but macro inefficient.
• If fair value cannot be determined it is harder for a market to be efficient.
• If there is no idea of what is fair value, new information will be misused.
• Our idea of market efficiency has evolved over the last few decades. 

A simple five-point investment framework checklist

Here is a simple checklist to compare different investment strategies. Call it a five-point smell test of what any investor should be looking for when a strategy is being pitched to them. 

• Strategy Design - A strategy should be easy to describe to investors. The underlying assumptions should be clear. There should be a strong economic or structural rationale for why the strategy should be successful. Some strategies will only be  effective for a limited time, so the manager should be able to explain the conditions when the strategy will fail.
• Expected returns - Defining the expected return goes without saying; however, expected returns should always be tempered by volatility. The Sharpe ratio should be strongly positive but also stable and consistent and reasonable. High risk adjusted returns will not last forever. 
• Reliability – All strategies are time varying. The reliability issue is whether it is clear when this variability will occur. What are the conditions for when an investment will do well or poorly? Back-tests are always good, but do they make sense relative to different market conditions.
• Convexity – Every investor wants convexity. It is always an issue of what is the price to be paid for this convexity. In a very broad sense, convexity relative to the market means that when the market is going up, the strategy will also go up and when the market goes down, the strategy will not lose as much as the market. Once that minimum achieved, a great convexity instrument will go up when the market goes down and will have positive correlation when the market is up. This is hard to achieve in reality, but the question has to be asked.
• Costs – There is the cost with implementing a strategy and there is the cost of downside risk. Both of these have to be reviewed and discussed. There is also a cost with liquidity if you are not allowed to exit a strategy or the price of exit is high.

Every strategy has a narrative. Having a formal checklist does not negate the narrative but provides a framework or lens for comparing investment narratives. 

Corporate spreads opportunities - Spikes and reversals

In spite of positive, albeit modest growth and firm equity prices, corporate spreads for lower quality corporates have started to move higher. BBB corporate spreads have come off lows for the year and are still below average. BB corporate spreads are also off their lows which have been tested twice before. A careful look at the quality spread of BB - BBB shows strong widening. We have seen this spread performance before with this behavior being the potential fifth quality spike over the last year. The charts show the BBB, BB, and the quality differential against three different  moving averages.

Corporate spreads as measured by the BAML indices represent the weighted average of yield above Treasuries across a rating. There will be dispersion across individual names and sectors. The spread includes the credit risk premium, the expected default risk, downgrade risk, and a liquidity premium. There will be a link with equity risks associated with the fundamentals of corporate finance. There will be a link with volatility and the flow of funds. These spreads will increase if there is an equity decline, a slowdown in economic growth, and a spike in volatility. While there are growing risks with the growth of debt and leverage, a true sustained spike in spreads may be premature. However, we have seen a number of quality spikes (BB-BBB) this year that have then reversed. These are technical opportunities for active traders. 

A framework for AI decision modeling

Everyone is talking about employing AI, but it is just a tool. It is a great new tool; however, tools are used to solve problems. If you apply the wrong tool to a problem, you will not get a good solution. To find the right tool, you have to define the problem. The framing of the problem is more important than the tools that are used to solve the problem.

The authors of Prediction Machines present a simple way to address problem framing called the AI Canvas. There are two parts to the canvas. One, framing the problem. Two, determining the data set needed and how to use it. The decision framing starts with a set of simple questions:

• The action - What are you trying to do?
• The prediction - What do you need to know to make a decision? 
• The judgment - How do you value different outcomes and errors?
• The outcome - what are the metrics for task success?

These four questions help the researcher determine the right tools, but to implement an AI program a second set of questions needs to be answered. 

• Input - What data do you need to run the predictive algorithm?
• Training - What do you need to train the algorithm?
• Feedback - How can you use the outcomes to improve the algorithm?  

The authors present this as a canvas because this framework can actually fit on a piece of paper.  Each box can be filled out to address the set of standard questions. If this is done carefully, it will take some time and effort before there is even a discussion of the techniques that could be employed. 

You may prefer a different framework, but the basic components should be the same: what is the objective, what data is needed, how will a model be tested, what output is needed for taking action, and how will outcomes be judged. Problem management does not start with the tools but with the framing of the problem. 

The "recourse against bad ideas"

"The only recourse we have against bad ideas is to resist the seduction of the 'obvious', be skeptical of promised miracles, question the evidence, be patient with complexity and honest about what we know and what we can know." 

- Abhijit Banerjee and Esther Duflo 

Nobel Prize winners in economics for 2019 

- Taken from The Economist book review of Good Economics for Hard Times  

They are a great team conducting careful economic development and microeconomic research. If every Wall Street analyst followed this prescription, we would have better quality research. Copy this quote and place it up on your wall if you are doing research. I cannot add to their advice.  

EY 2019 Global Alternative Fund Survey - Mature firms have grown-up issues

The 2019 EY Global Alternative Fund Survey provides some interesting reading on the state of the business for alternative funds. The survey has little to do with the current state of investing, but with how funds are managing their businesses. 

The key investment issue is that investors are tilting exposures away from hedge fund and toward higher returning private equity funds. The investor's search for yield and return continues and hedge funds have to not only compete against traditional asset managers and their peers but other alternative managers. 

The core survey business questions focus on strategic priorities, talent management, investments in data and technology, and the future landscape. The questions and answers suggest a more mature industry that is grappling with asset acquisition, talent management and cost containment. Investors also believe their focus should be on cost containment, talent management and succession planning. Innovation and improved return generation were not listed as strategic priorities.

Talent management priorities included retention, technology skills, and inclusiveness of culture. Data priorities included compliance, order/management enterprise systems, and fund accounting as critical issues. The future landscape question found passive investments as the big disruptor to alternative businesses. 

These business issues are not unique to alternative firms but seem to be core issues for any asset management or finance company. In fact, the survey seemed to be review of the key issues for any large corporation. Alternative fund firms have grown-up and face grown-up management problems. 

The drivers of equity return by decade - The winner for this decade: earnings growth

I was surprised by the this very useful table from Ben Carlson in his Fortune article "What Powered Such a Great Decade For Stocks? This Formula Explains It All". Looking at the drivers of equity return by data shows that the most recent period has been dominated by earnings growth not P/E expansion. Corporations have been making more money and that causes prices to rise. Earnings have trended up during this whole recovery and companies are currently beating expectations.

It could be a host of reasons for this earnings growth: lower costs of leverage, market power, cost containment, and taxes to name a few. But, there has been a limited increase in P/E ratios which suggest that the gains have not been from speculative excesses. There are clearly companies and industries that have high P/E's and P/E ratio are high on a historical basis as measured by Shiller, but it is hard to argue for extremes. If economic growth remains stable even if low, it is hard to argue for a market decline. 

"Kind" versus "wicked" learning environment - Financial markets are not kind

Good investment decisions focus on making good predictions or inferences. Improving these predictions is all about learning from the information available and past investment decisions. Unfortunately, learning is hard because the environment for making decisions is often complex. Most investors think it is hard because there has to be “sweat equity” to improve. That is false or at least only a part of the story. Learning issues and the path for improved decisions are associated with the type of environment faced. 

Learning is easy if there is a close well-defined link between action and results. There is a close link if the setting for gathering information is the same as the one faced when the decision is made. This learning problem is placed in a useful framework by one of the most eminent professors in decision science, Robin Hogarth, along with some of his collogues in the paper, “The Two Settings of Kind and Wicked Learning Environments”. (See Current Directions in Psychological Science vol 24 (5) 2015 pp.379-385.)

The paper breakdowns decisions into two settings, the learning or acquired information setting, and the applied or predictions setting. These two setting define the learning environment. The learning environment can be described as either being kind or wicked. If the learning environment is kind, then the information in the acquisition setting will be similar to information in the prediction setting. There can be accurate inferences because in simple terms, “what you see is what you get”. When the acquiring setting does not match the prediction setting, there will be mismatches between what is perceived and what is actually faced. This a wicked learning environment. What you think you know is not the reality faced.

 

Think of two simple decision environments, a well-defined game versus the investment decision game. A gaming environment with well-defined rules is a kind environment because mistakes can be defined, feedback given, and learning acquired. In the trading and investment game, players change, behaviors change, the economic structures and rules change. Given there is no specific end, it is hard to acquire feedback and effectively learn. The first environment is kind. The second environment is wicked.

In a kind learning environment, hard work will be properly rewarded. A player may not always be rewarded, but there is a close link between, information, decision, action, and feedback. This will not be the case with a wicked environment where in many cases the feedback received could wrong. The sample of information acquired may not be helpful for explaining the action taken.  Probability judgments in a kind environment means the favorable odds can be found and exploited. In a wicked environment, it is hard to distinguish between skill and luck.

A kind environment has plentiful and accurate feedback. Positive feedback will be harder to find in wicked environments, so mistakes are more likely. In a kind environment, statistical relationships from the past can be used to make accurate predictions on the future. Test or training sets will match prediction sets. There is a strong positive covariance between the past and the present.

Decisions can be placed within two settings, the learning and the target setting where the decisions are made. The framework can be effectively explained through Venn diagrams. When the learning and target settings closely match, the world is kind. Judgments are easy. When learning and target settings do not match, the environments are wicked, and judgments are harder to make and any improvement in decision-making is harder.

Creating the wrong learning setting through using bad information or forming biased use of information will mean predictions will be poor and there will be errors in judgments. Investors have to recognize the environment they are facing and what they can do to better match learning with predictions. First, realize that a given environment is wicked. Second, protect against bad judgments through focusing on the form of the feedback generated. Third, always work to improve your information setting through gathering more information or eliminating noisy information. Fourth, don’t confuse skill with luck. You are likely not as good as you think.