U.S. Equity Investor Letter for the Year-End 2023
By Haruki Toyama, Head of Large and Mid Cap Equity, Portfolio Manager
The U.S. stock market had a good year in 2023. The S&P 500 Index recovered all of its 2022 decline, and slightly more, ending the year up 26.3%.
The story of the year was the glaring concentration of returns in the Magnificent Seven, as the press has dubbed them. These seven megacap technology stocks returned an average of 111.7% for the year, accounting for well over half of the total return of the S&P 500. Until the last few months of the year, in fact, the index was essentially flat, excluding these seven stocks.
The broader story for the stock market in 2023 was that bigger was better across the board, keeping with much of the past decade, but more extreme than usual. Below are the total returns for the year, from the largest market capitalization index to the smallest.
Here are the same indices’ annualized returns for the past ten years, set alongside their returns for the previous ten years.
It’s not fully symmetrical, but it’s close. The reversals of fortune bring to mind the Horace quote Benjamin Graham chose as the epigraph for his magnum opus Security Analysis, “Many shall be restored that now are fallen and many shall fall that now are in honor.”
During the two-year stretch of 2020 and 2021, the Information Technology sector was the best performing sector within the S&P 500. That reversed in 2022, as the sector was one of the worst performing. And then, in 2023, it was once again the top performing sector. Honor, fall, restoration.
Whether it’s performance by market capitalization, sectors, or any other factor, stock markets are intrinsically cyclical. Some cycles are long-term, taking decades to unfold, and some are short-term, lasting months, weeks, or even days. Many are medium in length, lasting two, three, or several years. Most cycles occur because a trend often creates the seeds of its own reversal. We at Madison Investments are certain that market cycles will occur, but it doesn’t mean we can predict their timing or magnitude. We don’t think we can. This is perhaps a major difference between us and many other investors. Most investors believe it’s their job to time market cycles despite overwhelming evidence that it’s nearly impossible to do so with enough accuracy to make such an effort profitable over long periods. We avoid making calls about market cycles and spend zero minutes thinking about them, not because we don’t think they can be important, but because we think they’re inherently unpredictable in duration.
This mentality of our team is generally true for other kinds of cycles, such as macroeconomic, industry, or company-specific, but is a bit more nuanced for those. We make no explicit prediction about cycles on which we base a buy or sell decision. Still, we are acutely aware of the various cyclical forces at work, and depending on whether we think we have the ability to assess the length or intensity of such, we may incorporate them to various degrees.
Let’s use a few examples to illustrate our point. We’ve been invested in off-price retailer TJX Companies for just under ten years, having invested in 2014 in our Large Cap strategy. TJX is one of the most recession-resistant companies we own due to its perennial value proposition to customers; customers always like to save money, especially when economic times get tough. As a result, the company has had an exceedingly steady revenue and earning profile over the past several decades.
But that doesn’t make it immune to other kinds of cycles. After our initial investment, the stock materially outperformed the S&P 500 for two years. Then, it materially underperformed for two years. Then, it materially outperformed again for two years. Then, once again, it materially underperformed for over two years before starting its current stretch of strong two-year outperformance. Despite these swings in relative performance, the investment has done rather well over the full term of our ownership, outperforming the index.
For some periods of underperformance, there seem to be fundamental explanations – the company made a few merchandising mistakes, and sales were weak, or there were expense pressures that crimped margins, for example. For the other periods of underperformance, there are no obvious reasons for underperformance. The stock likely just fell out of favor with investors for market cycle reasons or perhaps “got ahead of itself.” But the critical point is that we made little attempt to forecast these unforecastables, and held steady with our investment with infrequent trading. So far we have been rewarded.
Let’s take another example of a recession-resistant investment we’ve held for many years, Brown & Brown. We first purchased this company in 2007 in our Mid Cap strategy. As an insurance broker, it gets paid a commission on the premiums that its mostly small business clients pay. Since clients need to maintain insurance coverage even in business downturns, Brown & Brown’s revenues tend to be very steady year by year. Yet, our investment underperformed for the seven years after our initial purchase, and it wasn’t because we paid a high price – the stock traded at a moderate price to earnings (P/E) of 17x at the time. The culprit was profits. After increasing sixfold over the seven years before our purchase, earnings per share were essentially flat from 2007 to 2014, going from $0.68 per share to $0.71 per share. No wonder our investment underperformed the Russell Midcap benchmark over that period. The sources of sluggish profits were manifold, including management turnover, a change in its acquisition strategy, moderate under-investments in dealing with the shift towards more complex insurance needs among its customer base, and a heavy exposure to Florida, a state hit especially hard during the Great Financial Crisis.
Perhaps some of these factors could have been foreseen in 2007, it’s hard to say. What we can say, though, is that by 2014, we had the full explanation of these factors in hand, and our continuing research indicated that many of these factors were being addressed or were likely to diminish in importance. The exposure to Florida had declined, and the negative impact of the Great Financial Crisis on the state’s economic growth was receding. Management departures were stemmed, and we had full confidence in the CEO, who was a long-term employee and an owner-operator willing to make changes to address long-term health. One of them was to make the investments needed to keep up with the increasingly complex insurance needs of its clients. And finally, we felt that its acquisition strategy was a change, but a positive and necessary one. Because of these assessments, we felt that the future looked bright, much brighter than the recent past, and we added significantly to our investment during 2014 and 2015. We turned out to be right. The company is on track to produce close to $3 per share in earnings in 2023; as a result, its stock price has followed suit with strong performance as well, to the point that our investment has now outperformed the benchmark since our original purchase, despite the many years of underperformance initially. This is an example of a long-term earnings cycle based on microeconomic factors, mostly internal to the company itself. We feel that our research style is well-equipped to analyze these kinds of cycles.
These two examples illustrate why we much prefer to invest in companies less exposed to the vicissitudes of economic and industry cycles. Such a philosophy takes one factor out of the equation when evaluating an investment, one where we don’t think we have any edge. The success (or failure) of an investment is dependent much more on the quality and depth of our research and insight, rather than on outside forces that we don’t control and have little to no ability to predict.
However, we don’t entirely avoid cyclical companies if we feel the valuation is compelling and the long-term outlook is sufficiently attractive.
An example of that would be Moelis & Co, a boutique investment bank that specializes in mergers and acquisitions (M&A) and restructuring advice. M&A is a notoriously cyclical industry, similar to another industry we will discuss, semiconductors. Much like semiconductors, it has a history of secular growth and prospects for similar growth in the future appear good as well. In addition, we believe that Moelis has an organizational culture that makes it almost unique in the M&A advisory business – in the infamously individualistic industry, it has added a healthy dollop of teamwork ethos. In a business where your primary assets walk out the door every evening to go home, this is a crucial feature for the sustainability of your franchise.
When we invested in mid-2020, Moelis’s revenues were under pressure from the pandemic that had virtually shut down merger activity, and the company was losing money. Yet, we knew that at some point, M&A activity would resume, and we also could see that the company had the balance sheet to withstand a protracted period of low M&A activity. The stock traded for what we calculated as a single-digit P/E on future earning power. In contrast to our recent semiconductor investments, we were fortunate on the timing, and the industry recovered shortly after. While we believe our insights into the specific qualities and strengths of Moelis have been a meaningful contributing factor to the investment’s success thus far, a larger contribution can perhaps be attributed to our willingness to look through the downturn, and not attempt to make a precise prediction on when a recovery would occur. When J.P. Morgan was once asked whether he prognosticated a recession or an economic boom, he answered simply, “yes.” That’s how we valued Moelis.
Another example of a venture into a cyclical industry, but one which has not worked out for us thus far, is our investment in four semiconductor companies, accumulated over the past several years, much of it over the past two.
In the semiconductor industry, demand is closely tied to the economic cycle, and industry order patterns are magnified by its complex supply chain. Furthermore, it’s a capital and technology intensive industry, resulting in frequent overcapacity and product obsolescence. For these reasons, we shied away from any investment in the industry for many years. However, several years ago, we felt that segments of the industry had developed strong enough moats to merit investment under the right conditions. We made one investment (Analog Devices in Large Cap) that we thought was compelling, but it wasn’t until late 2021 when the semiconductor industry entered a severe downturn, that we began purchasing semiconductor stocks in earnest. Our view on the cycle risk was simple – the long-term growth outlook was tremendous given the proliferation of electronics and software in everyday objects, and the downturn had already begun, and a healthy amount of it was known. We ignored the near-term outlook for profits, assuming they would be down substantially. It turned out that the downturn was more severe than most other investors or industry observers had forecasted. Thus, semiconductor stocks in general remain depressed, and our investments have so far shown mixed results.
However, we have not changed our view on the long-term attractiveness of the industry segments in which they operate, nor on the normalized earning power of these companies five to ten years out. Thus, we have been consistently adding to our exposure for the past two years, which, in addition to Analog Devices, now includes Microchip Technology (Mid Cap), MKS Instruments (Mid Cap), and Texas Instruments (Large Cap), the last of which was a new purchase in 2023. Each of these has wide and deep moats, top-notch management teams, and strong secular growth prospects. We expect our investments here to ultimately work out for us, but they are examples of how unpredictable, cyclical factors can overwhelm other factors in the short term.
We can use the examples above to distill our working philosophy when it comes to cycles.
We ignore stock market cycles. We have never made an investment decision because we thought that “megacaps are poised for a comeuppance,” or “the consumer sector is going to outperform,” or “high beta stocks will do well this year,” or any other such nonsense.
We mostly ignore macroeconomic cycles. We pay attention to them only because we know downturns will inevitably occur, and will also inevitably end. Thus, we incorporate them into our estimation of normalized long-term earnings for companies exposed to such cycles, and thus into our valuations and our required margin of safety. If we feel that certain key macroeconomic factors have undue influence on the long-term earning power of a company, we take a pass and move on.
We pay heavy attention to industry and company-specific cycles. This is our value-add as investors. Our research revolves around distinguishing that which is temporary and that which is permanent. If we can’t tell the difference when researching a prospective investment, we move on.
We demand above-average long-term profit growth. We are fine investing in cyclical companies, but they must have the capacity to grow earnings over successive cycles. As long as we’re comfortable that a company will have much higher earning power many years out, we should be able to ride out any downturns in the interim.
These guidelines work for us, because of our long-term orientation and the long-term orientation of our clients. We have owned the average stock in our Large Cap and Mid Cap portfolios for seven and a half years. Over a third of the combined portfolio is invested in companies we’ve owned for over a decade. This would not be possible if we were judged only on short-term performance. For that, we are grateful.
Respectfully,
Haruki Toyama
View previous letters from Haruki Toyama: