This post is a continuation of a series where we will be providing some of our own thoughts and commentary on the conversations we had in the first season of our new podcast.

This post covers our conversation with JD Gardner, which you can listen to here.

 

2:32 – Corey and JD kick off the podcast with a bit of basketball talk.

Nathan Faber (“NF”): Having grown up in Ohio, you might think my natural tendency would be to pick Lebron… but I agree with Jordan.

Justin Sibears (“JS”): For me, it’s all about the rings.  Jordan is 6-0 in the finals.  Lebron is 3-6.  Advantage Jordan?  Not so fast. Since we are quants, we are obligated to dig more deeply into the numbers.  Fortunately, FiveThirtyEight has done the work for us. Lebron has actually slightly overachieved in terms of rings won when adjusting for the quality of his teammates and the opposition.  FiveThirtyEight calculates Lebron as having 2.8 expected finals wins vs. his 3.0 actual. Alas, Jordan wins even when we use more complicated math, with 6.0 actual wins relative to 4.1 expected.

4:02 – JD discusses his background, playing basketball in college, and taking the CFA.

6:14 – The penny stock trade that financed JD’s wife’s wedding ring.

NF: JD’s assessment of 100% luck on that trade helped him not double down and lose it. He put the winnings to good use!

Corey Hoffstein (“CH”): Normally when investors are seduced by penny stocks early in their career, it ends in a horror story.  JD is the rare exception and he was fortunate to take his winnings and bail.

7:34 – JD opens up about his first job as a financial advisor’s assistant and the behavioral foibles he witnessed among clients.

NF: Seeing behavioral biases not only in clients but also in advisors shows the need for a systematic process. Having a client’s and an advisor’s behavioral biases coincide is a recipe for disaster.

JS: There are two general approaches to controlling for behavioral biases.  One is to change the investment returns (e.g. trying to deliver returns that are less likely to cause investors to make poor decisions).  The other is to change the investor (e.g. trying to influence them to make better decisions).  Here is a discussion of this topic by Betterment’s Dan Egan (based on a question posed by Corey).  Realistically, I think both elements are important.  Short of a portfolio that never goes down and always beats the market, it’s likely too much to ask for the portfolio alone to eliminate behavioral biases.

CH: The theme of this season, for me, has been that experiences ultimately have a large impact on the way people manage money.  One of JD’s earliest experiences was witnessing the behavioral foibles of individual investors and the struggles related to keeping them on plan.  As he discusses later, this experience has a big impact on the types of strategies he thinks about bringing to market at Aptus.

11:22 – Lessons that JD learned in trying to control and guide client misbehavior.

NF: JD discusses conviction and process. I think that it is important to highlight the need for conviction inprocess. If you don’t believe in the system, how can you stick with it during the trying times? I know that Aptus tries to reduce the magnitude of losses during market environments that are rough for their value strategy. This should make conviction easier as long as the premiums paid in positive market environments are not too high.

12:31 – JD discusses his transition into a deep value analyst role, but expresses frustration in his ability to communicate expectations to clients.

NF: It is hard enough to set expectations about a diversified portfolio, much less a single deep value stock.

CH: JD echoes a sentiment that I’ve heard most articulately put by our friend Wes Gray over at Alpha Architect: “sustainable alpha requires a sustainable investor.”  Investor (mis-)behavior can turn a great long-term strategy into horrible realized results.

16:18 – JD is introduced to a world almost completely opposite to his experience in deep value research: managed futures.

NF: I am sure this lesson in removing subjectivity and having a strict set of rules really informed JD’s current philosophy.

CH: I am a big proponent of learning about other ways of investing, even if you do not apply them directly to what you can do.  Often ideas seen in a different context can seed creativity within your own field.  While the distance between deep value research on individual equities and systematic, managed futures may seem vast, we’ll hear later that both of these approaches had a significant impact on the way JD thinks about building portfolios today.

18:55 – Finding harmony between fundamental research and systematic investing.

NF: A lot of fundamental research takes a long time to pan out. However, even systematic strategies will go through periods of short term underperformance.

JS: One of the very underrated aspects of systematic investing is that it forces you to make important strategy design decisions upfront (setting aside having the flexibility to amend the models over time).  In my experience, this gives you the opportunity to thoughtfully consider the impact of these decisions (i.e. number of holdings, how holdings are weighted, how frequently are rebalances implemented, etc.).

CH: My philosophy is that every discretionary manager can be broken down into two component strategies: their purely systematic approach and the idiosyncratic decisions they make.  This framework leads to two takeaways.

First, it means that for us to elect a discretionary manager over a systematic approach, we need to develop confidence in the discretionary manager’s ability to make idiosyncratic decisions.  Or, similarly, we may wish to operate in a space that cannot be easily systematized (e.g. illiquid debt or event-driven strategies).

Second, it highlights the trade-off in purely systematic versus discretionary.  In a purely systematic approach, biases are embedded upfront and held consistent.  In a discretionary approach, those same biases may exist, but a discretionary manager has the ability to course-correct.  On the other hand, it also means that the discretionary manager may be subject to ongoing behavioral biases.

My view is that if biases are going to be embedded either way, I would prefer the systematic approach: at least then I can try to be aware of the biases I am exposing myself to.

20:21 – Exploring the philosophy of “winning bigger than you lose.”

NF: This may seem obvious, but it is easier to implement than trying to always win.

CH: In a simplified framework, there are really two levers to try to pull in designing a strategy: accuracy and the size of winners versus losers.  The holy grail, of course, is a strategy that is both highly accurate and where the size of gains far exceeds the size of losers.  In practice, these factors tend to exist in balance with one other.  Convergent strategies tend to have many small winners, but a few large losers (e.g. writing puts or merger arbitrage) whereas divergent strategies tend to have a lot of small losers but a few big winners (e.g. managed futures).  Most strategies exist within these extremes, but understanding which lever we’re trying to pull can be very important for understanding the behavior of our system.

22:21 – What’s more important: the relative size of your winners versus losers or your overall accuracy?

NF: My take on this is that both can be equally important from a theoretical standpoint, but the behavioral aspect is likely easier when winners are much bigger than the losers. Winning more frequently does not erase the larger pain investors feel from having big losers.

JS: I have a slightly different take on this than Nathan.  Yes, mathematically the relative size of your winners vs. losers and accuracy are equally important. A strategy that has a 90% chance of returning +10% and a 10% chance of returning -10% has the same expected return (8%) as a strategy that has a 10% change of returning +170% and a 90% change of returning -10%.  However, I think the size of winners and especially losers can be more important from a risk management perspective since catastrophic losers can be nearly impossible to recover from.  In addition, the focus on the size of winners and losers forces thought to be put into position sizing, drawdown control, etc. that may help protect the portfolio from a downside event than is even worse that anything we’ve seen historically.

CH: I agree with Justin; the multi-period nature of investing can mean that a single catastrophic loss can wipe out a portfolio.  Of course, so can the compounding of a large number of small losses…

23:55 – “Diversification is the most overused word in the financial world.”

NF: We have looked at this in the past. In our commentary, Is your Multi-Asset Really Multi-Asset, we explored the dangers of assuming that a strategy with different asset classes is truly diversified from a risk factor point-of-view.

JS: Strategy diversification is very underutilized in our experience.

26:10 – Where does Aptus fit in the landscape of value investors?

CH:  What I found most interesting about JD’s answer was that it really was less about value investing and more about the value investor.  He acknowledges that his strategy is designed for retail investors who have more limited investment horizons that institutions, and therefore he has to implement a form of value than is achievable for that audience.

27:41 – JD explains the basics of the Fortified Value strategy.

NF: Combing value/quality factors with an equity tail hedge is an interesting way to try to offset the premiums from the tail hedge in positive markets.

JS: I often think that value can be a particularly hard strategy to stick with because its contrarian nature can be stomach churning.  Hearing a portfolio manager tell you that even though this stock is down 50% this year and it has lost money the last six quarters, it’s a great long-term buy because the liquidation value of its assets exceeds enterprise value can cause some heartburn.  Incorporating a quality component makes the story a bit more warm and fuzzy.  Buying undervalued companies with strong profitability and growth just is more comfortable.

28:31 – The origin of the idea for Fortified Value.

CH: I think the most interesting comments here – and mentioned later in the conversation by JD – is the idea of launching two strategies that are designed to complement one another.  Often strategies are evaluated on a stand-alone basis.  If we knew, however, that a strategy was always going to be held alongside another existing strategy, some interesting design choices arise.

29:31 – The Fortified Value process and building a more “achievable” value strategy.

JS: Many composite measures of a factor, like value, simply combine a number of different, yet similar, measures (e.g. price-to-earnings, price-to-sales, price-to-book).  I find it interesting that JD’s three metrics each seem to serve a different purpose. ROIC is more of a quality metric. Cash flow to enterprise value is where the core value exposure comes into play.  P/E relative to historical P/E for that stock tries to address stocks like Amazon that can be difficult to value.

CH: The use of P/E versus historical P/E is interesting because it allows the strategy to incorporate highly valued growth companies that may be selling cheaply versus their own historical value, but still be expensive relative to the market. In many ways, this might tilt the strategy to somewhat of a “growth at a recently-more-reasonable price” strategy.

32:45 – Is this a diluted form of value investing?

NF: It all depends on how you define the value premium. If a company has room to run relative to its own history, that can be “value”, even if it is not in the traditional sense. An issue can arise for investors when comparing a strategy like this to other value strategies. There will be tracking error, but it does not mean that looking at different measures of value is wrong.

CH: Traditional quantitative value buys things that are relatively cheap, measured through a variety of metrics.  However, a company can have seemingly expensive metrics – like P/E – that are entirely justified by growth rates.  Indeed, a high P/E can still be cheap if the growth rate is higher than expected by the market.  I sort of think of the use of P/E versus historical P/E as the assumption that the market is generally efficient at pricing companies, but they can go through short-term mis-pricings.  This logic would have us believe that a low P/E is justified as much as a high P/E is justified; but a P/E lower than historical average might represent a value opportunity, regardless of the absolute level.

34:05 – How do you build a portfolio around the idea of “winning bigger than you lose?”  Let winners run.

NF: Many value strategies (e.g. smart beta ETFs) rebalance annually, but the value premium sometimes takes longer to mature. If you have a stock with good momentum that has just moved above the threshold to be in the portfolio based on whatever metric you use, there can be an opportunity to capture a higher premium by holding a little longer. This is an aspect of the diversification that value and momentum have with each other and is more of an integrated approach to combining the two factors.

JS: There are a number of simple ways to implement something like this.  One is to explicitly build momentum into the equation. For example, never sell a holding that is in the top 20% of the universe by trailing 12-month return.  Another is to have different triggers for buying and selling.  For example, buy stocks that are in the bottom 10% by price-to-earnings, hold until stocks move into the top 50%.

38:51 – The behavioral benefit of looking less “suicidally value.”

NF: The risk of catching a falling knife can be a trigger to violate the value system.

CH: “A strategy’s return is much less important than an investor’s return while exposed to the strategy.”  I think this statement perfectly sums up the JD’s entire founding thesis.

41:28 – Exploring the put option protection embedded in the Fortified Value portfolio.

NF: Dynamically allocating to puts is interesting. The issue is that when volatility spikes, buying puts at a fixed OTM percentage can be more expensive.

JS: We recently published a piece on the benefits of diversifying across multiple risk management approaches.

CH: What I found most interesting about this approach is that by using a constant proportion of capital to buy the puts, the strategy will end up buying more options when volatility is low and fewer options when volatility is high.  This is in contrast to an approach that tries to protect a fixed proportion of the portfolio regardless of the market’s behavior and can end up spending a significant amount of capital in high volatility environments.

45:35 – Corey asks JD to defend the use of put options where research suggests that put options provide little true protection for their cost.  Corey specifically references the paper Pathetic Protection: The Elusive Benefits of Protective Puts.

NF: One key difference between JD’s index and the AQR paper is that the AQR paper looked at puts priced at 5% OTM. When JD is buying puts 25-30% OTM, the detriment of buying the puts may be different than that seen in the AQR paper. However, with a higher volatility premium on the deeper OTM puts, I would expect the paper results to hold.

CH: While I think the AQR paper brings up many great points – such as the path-dependent risks of rolling put options – I will come to JD’s defense and point out that the CBOE PPUT index referenced in the paper buys a single 5% OTM put option each month.  As volatility regimes fluctuate, this option will represent a different proportion of capital spend.  In highly volatile environments, it might represent far more than 0.5% of the portfolio’s notional value.

47:54 – JD outlines how they implement the put option strategy in a valuation-dependent manner.

NF: Being dynamic with the put allocation based on expected drawdowns may be what it takes to make the protection worthwhile. But to play devil’s advocate, if the purpose is to protect from the Black Swan events, it seems like you would always need some protection. I would be interested to see how the dynamic put exposure would have behaved in a 1987-like environment.

JS: This section is interesting.  In a sense, the tail hedging strategy is a bit less about using options and a bit more about “time-series value,” making tactical decisions based on whether the market is under or overvalued vs. history.  In this way, Fortified Value is cross-sectional value + time-series value while Behavioral Momentum is cross-sectional momentum + time-series momentum (also known as trend).

49:18 – Thinking of the put options less as protection and more as generating capital that can be invested when markets fall.

NF: This is a totally new mental model for me. Holding the puts for the purpose of investing more capital in value stocks at the cheap times is very intriguing.

CH: I agree with Nathan; I thought this was a really unique idea.  The market aphorism of, “buy when there is blood in the streets,” requires us having available capital to buy!  The “hedge” represents an intriguing way to generate capital we can put to work if we think securities are temporarily mispriced in a market dislocation.

51:13 – If you were an investment strategy, what would you be and why?

 

Corey is co-founder and Chief Investment Officer of Newfound Research, a quantitative asset manager offering a suite of separately managed accounts and mutual funds. At Newfound, Corey is responsible for portfolio management, investment research, strategy development, and communication of the firm's views to clients.

Prior to offering asset management services, Newfound licensed research from the quantitative investment models developed by Corey. At peak, this research helped steer the tactical allocation decisions for upwards of $10bn.

Corey is a frequent speaker on industry panels and contributes to ETF.com, ETF Trends, and Forbes.com’s Great Speculations blog. He was named a 2014 ETF All Star by ETF.com.

Corey holds a Master of Science in Computational Finance from Carnegie Mellon University and a Bachelor of Science in Computer Science, cum laude, from Cornell University.

You can connect with Corey on LinkedIn or Twitter.

Or schedule a time to connect.