Q: What is the history of the fund?
The Columbia Small Cap Value Fund is a small-cap portfolio that was launched in July 1986.
Since I took over as manager in 2002, the fund has had a philosophy and process based on what we call “the trinity,” or three key attributes of companies in our investible universe that create a strong margin of safety and increase their probability of success.
Our investment process has evolved when presented with opportunities to improve, as it did five years ago after the fund experienced its only significant shortfall in absolute performance over the past 16 years.
In 2012, the 10-Year U.S. Treasury plunged. With this sudden and large decrease in the average cost of capital, lower-quality companies could seemingly create earnings by refinancing debt – something inconsistent with the philosophy of our high-quality, small-cap value portfolio.
We tapped into the vast resources at Columbia to better understand the market dynamics, and in doing so, came away with an important lesson: with 190 to 210 names, the portfolio was and is quite diversified. But because every name is consistent with our high-quality philosophy, if our philosophy underperforms, so will the fund – and in 2012, it did.
So, we took a deep dive into quality itself. When we peeled back the onion, we found two characteristics that are orthogonal to high quality: momentum and growth. Of the two, we believed growth was most consistent with a small-cap value portfolio like ours, so long as we were paying below intrinsic value for that growth.
Based on these lessons, and without straying from our core mandate, we made enhancements to the portfolio and the process by adding a growth vector. Although it sounds like a big hurdle, we now have another tool in the toolbox to help us try to obtain at least a 12% discount to intrinsic value.
The proof point of this evolution came in 2016 when the 10-year U.S. bond yields plunged again. Our performance was significantly better than our peers and benchmark, and we believe adding a growth vector has diversified the portfolio overall.
Q: What core principles drive your investment philosophy?
By centering on our trinity, we identify attractive, small-cap value companies trading at a discount to intrinsic value. When invested for a full market cycle, we believe these companies will tend to outperform.
Our core trinity is similar to a time-tested, Graham and Dodd approach. The first of its three parts focuses on high-quality balance sheets, the second on strong cash flows, and the third on discount to intrinsic value. The modifier we added five years ago identifies companies with good growth prospects.
Often, we see an opportunity set where others do not, because our fundamental understanding of high quality is different. When investors look at balance sheets, many define quality as a high return on invested capital (ROIC) or a high return on equity (ROE).
They frequently make the mistake that low ROEs are a hallmark of poor-quality companies. We have found this metric sometimes indicates exactly the opposite; many fine companies with strong balance sheets have low ROEs because they have been artificially depressed by an excess of equity.
We look for companies with strong balance sheets but low ROEs because they are often discounted in the market. Our fundamental analysis helps us understand how their excess equity could be employed for our clients’ benefit.
Q: What is your investment process?
Because the small-cap value universe we focus on is large and dynamic with upward of 1,500 names, wading through it takes a number of steps.
First, our quantitative research team puts together a statistical model that shows us what is cheap or expensive in the market, which provides us a great starting point to begin our fundamental research.
Even though I am the portfolio manager, I still spend a tremendous amount of time with companies. On average over the last four years, I have met with 500 companies annually. Also, I cannot give enough credit to the 23 senior research analysts and seven junior analysts of Columbia Threadneedle who have a strong understanding of the competitive landscape. We also lean on the guidance of industry experts.
Our focus is on remaining disciplined to our process and philosophy over a three- to five-year time horizon. We do not waste time on companies unlikely to meet our criteria, and will pass on those with poor balance sheets, poorly structured balance sheets, or that do not generate cash flow.
Q: Would you illustrate your research process with the help of an example?
American Equity Investment Life Holding, which sells annuities and life insurance products, is a company we have a long history with.
Last year, it was trading in the low-to-mid 20s. The stock had moved down, then back up, and there was concern about the energy bonds in its portfolio. We continued following the company; its indexed annuities were a growing part of the financial tools we needed for our investors, and although we felt American Equity Life was a good niche product, it became too expensive for us.
In April 2016, when the Department of Labor (DOL) rule came out, people had expected indexed annuities would be excluded. When they were included, the stock of American Equity Life fell 60% that day.
We quickly huddled together with our insurance analyst. Because of our familiarity with the company, we knew its balance sheet and cash flows were strong and that it had a good growth vector, so the main question for us was whether there was a sufficient discount to intrinsic value. Also, given the concerns over the company’s energy portfolio, we analyzed the bonds and determined there were not significant default risks.
After just a few hours of analysis, we ended up purchasing the company for around $12.85. Having access to outstanding analysts and capable teams gives us the fundamental horsepower needed to take advantage of dislocations like this in the market. Currently, the stock is trading in the mid 20s.
Q: What influences your buy-and-sell decisions?
Intrinsic value is not a static number, so our decisions are not based on target prices. We do, however, set intrinsic value targets.
We continually assess the intrinsic value of our companies, making sure there is sufficient enough discount to intrinsic value to continue to hold. That 12% discount is always our guide post, and we generally rely on dividends for a safe and secure portion of that risk-adjusted return. To determine whether an underlying security has enough potential capital appreciation, we use price limits with probabilities around them.
There are three reasons for us to sell securities. The first is if a stock has appreciated to a fair value and no longer presents any significant upside.
Second, because the fund is pure small-cap value, we sell when stocks change their characteristics, for instance, if they appreciate past the upper end of the Russell 2000 Value Index, which is our benchmark, or if they become more of a growth play than small-cap.
Third, we sell when we were wrong. Sometimes things change – in the market, the competitive landscape, or in a company strategy – and are no longer congruent with our mandate. When this happens, we reassess the situation and sell, hopefully at a gain for our clients, but maybe not meeting our total return threshold.
Q: What are some important considerations for you in portfolio construction?
We manage 200 names in our portfolio, and being completely consistent with its construction is critical to us.
Looking across various small-cap value funds, they tend not to be able to maintain performance as their assets under management (AUM) grow. Instead, growth comes by adding name count or by migrating up market capitalization toward more liquid mid-cap names.
In 2002 when we took over this strategy, we made a pledge to our CIO we would not do this, that the fund would remain disciplined and invest in same kinds of companies until 2017 and beyond – and today, our AUM are approximately $1.2 billion.
We invest across the whole small-cap spectrum. The smallest name in the portfolio has a $90-million market cap and the largest is a company with a $5-billion market cap.
Our portfolio is diverse, and we want to make each of its 200 names count. I think of it as if it were a baseball team, and each of our holdings is a baseball player. Every day between 9:30 a.m. and 4:00 p.m., each player has to go out and take a swing. If a portfolio is weighted incorrectly, only the swings of a couple of those batters actually matter.
This is important to us when thinking about portfolio construction. If we mismanage weightings, only a precious few names actually determine the whole portfolio.
So, we use our own weighting scheme to ensure every name in the portfolio is contributing to active risk and to total return. It considers the liquidity, volatility, and potential upside of each security.
The biggest problem in portfolio construction occurs when people bet only on their highest conviction ideas; many people do not understand the risks and ramifications of doing so.
Sometimes our highest conviction ideas – like biotechnology companies – are the lowest weight in the portfolio, while some of our lowest conviction ideas – like utilities – tend to be our highest weights.
Why? Because the potential volatility of a utility company is significantly lower than that of a biotech. So holding 30 basis points of a biotech company might express far more conviction than a utility representing 1.2% of a portfolio.
We also use 13 thematic baskets when constructing the portfolio. They represent long-term tailwinds we have identified to the U.S. economy. These are not fads, but are tailwinds like water scarcity, the Internet of Things, infrastructure, aging population, and agriculture.
These themes are used in our portfolio construction because although we think the classic GICS sector view is helpful, it is also flawed. Our themes tend to run across a number of sectors, so using them gives us a better holistic view of what our clients are being exposed to. For example, when we talk with a company dealing with one aspect of infrastructure, that information can be used to increase our understanding of all companies with exposure to that same theme.
While our construction is benchmark aware, we are by no means benchmark dependent. Currently, 27% of our names are not in the benchmark.
Over long market cycles, turnover has remained between 25% and 40%. But we are quite flexible, and when times change, so do we. We see market dislocations as opportunities to shift the portfolio.
Last year, we repositioned the portfolio after seeing a significant risk due to the valuation of sectors like utilities, and lowered our weight there. Toward the end of the year, we significantly reallocated weight toward healthcare, an area we had been underweight for almost 18 months.
The portfolio’s overall turnover has a weighted average holding period of four years. Names in diversified financials are held, on average, for eight years; insurance companies, seven years; and banks, 5.5 years.
Compare those to areas we are excited about today because of dislocations in the market – areas like energy, where our average holding period is 1.6 years, and healthcare, where it is just 0.7 years.
Q: How do you define and manage risk?
Risk is a statistical term. In terms of portfolio management, the greatest risk is a permanent loss of capital, and we believe our philosophy minimizes the potential of this.
Correlations and concentrations can also contribute risk. In 2012, we had one significant concentration: all the names in the portfolio were consistent with our philosophy, which underperformed. However, we now understand how to diversify better in the future.
How names interact with each other is another risk consideration. A “what if” tool allows us to propose scenarios to see how the overall composition of risk in the portfolio changes as we enter names into it.
We have managed through two market cycles since we took over the strategy in 2002, and every single cycle presents another opportunity to learn.
For instance, in 2003, the benchmark dropped something like 50% that year, and that was coming off an already low level. We decomposed what hurt us; that year it was not our philosophy, but low-price risk. There had been such a sell-off in 2001-2002 that a number of names meeting our criteria were trading at $1 to $10, and most asset managers shy away from such low prices.
We learned from that, and used our new understanding to adjust the portfolio to make sure it would not have that risk moving forward.
In the years to come, we will identify further areas to evolve and enhance the strategy. It is only through truly understanding the portfolio and its risks that we can improve the process. Although we will probably continue to make mistakes, we will work hard to not make the same one twice.
