Dear friends and investors,
For the quarter ending March 31, 2020 Forage Capital (“the Fund”) returned an estimated -14.1%, net of all fees and expenses compared to -19.6% for the S&P 500 Total Return Index. At quarter-end, the portfolio was comprised of 15 companies and cash made up 10% of assets.
I have two theories about investor behavior. The first, informed by my observations in 2008/2009 and March 2020, is that market crashes turn fundamental analysts into macro prognosticators. Ensconced in the protective bubble of a rising market, an investor pontificates about moats and market opportunity. Then crisis strikes and such high-minded talk dissipates as rapidly as the unrealized gains substantiating it, replaced with high conviction views on sweeping macro themes. Rather than deliberate “at what price should I own this business?”, investors wonder “when is the right time to get into the market?” Those who could not wait to buy a stock in the event of a 10% drawdown when markets were ascending suddenly remark “it still feels too early” in the throes of a dramatic market decline.
The second theory is that when what makes sense runs counter to what has recently worked, investors will typically choose the latter, which causes them to learn the wrong lessons. In 2019, 10 years into the bull market – when in retrospect there was no dip that shouldn’t have been bought, when every sell decision looked like a mistake – there seemed broad support around the notion that one should never sell a competitively advantaged business, no matter how lofty the valuation. Compounding is so powerful a force (cue the Einstein quote!) that it compensates for whatever exorbitant price one might pay. Then this past month, with beloved compounders selling off with the rest of the market, some investors suddenly eschewed the “hold good businesses forever” dogma, congratulating themselves on panic selling and missing the last 15% (or whatever) of the market decline, confident of being just as prescient on the way up.
In my opinion, market crashes don’t generate new lessons as much as they reinforce ones that we should already know: that believing in and acting upon one’s private intuitions about the near-term market impact of complex macro phenomena is folly; that in being unable to reliably predict such things, we should own companies with resilient business models, durable competitive advantages, and strong balance sheets that can survive significant shocks; that we should buy those companies at prices that impute returns that meet our personal requirements.
This doesn’t mean you blithely ignore the COVID data. It just means you use it to update your judgment about each company’s long-term business prospects and near-term ability to survive rather than to formulate grandiose opinions about how the market will digest COVID headlines over the next 6 months (i.e. “there’s still more pain ahead and I just know that I’ll have a chance to buy this stock I like even cheaper”). I believe the latter is a fool’s game, one that becomes increasingly tempting to play during violent downturns, when it seems things can only get worse and when one’s personal opinions about unknowable market outcomes can perversely “feel” more valid.
Let’s just accept that the next several quarters (and maybe this whole year) will probably be brutal, even for many excellent companies. Assuming one is willing and able to look out 5+ years, the COVID-related questions that matter are: 1/ does the company have the balance sheet and cost structure to survive a severe demand contraction? and 2/ does COVID compromise the company’s competitive advantage or constrain its market opportunity? Additions/subtractions to what we own will reflect my updated relative degree of confidence in the answers to those questions.
For the moaty/growthy stuff that I typically follow, the first 2-3 years of cash flows comprise maybe 15%-20% of a company’s intrinsic value (if that). Even if COVID related demand shocks were to cut those cumulative cash flows in half, the resulting impairment in value would still not justify the 30% to 50% declines experienced thus far. Of course, before a company realizes its terminal value, it must first survive. I thought the latter criteria was met for a handful of companies that I added to the portfolio in March: Align Technology (ALGN), Booking Holdings (BKNG), Moody’s (MCO), HEICO (HEI/A), and SmartSheet (SMAR – starter position). I also bought shares in Credit Acceptance Corp. (CACC) in January (alas, too soon!), before Corona fears reached their present escalated state.
We own some stocks that are very much in Corona’s blast radius (position sizes as of 4/3/20):
Wizz Airlines (4% position): While legacy US carriers plowed their cash flows into share buybacks over the last few years and today find themselves in a badly compromised state, Wizz, understanding that a high fixed cost structure and negative revenue growth can be a lethal combination, stockpiled cash to survive the inevitable downturn. At the end of 2019, the company sat on €1.3bn of cash (~40% of its market cap) and had no financial debt. I think that the company could withstand an entire year of 0 revenue and still survive, even with limited cost cuts and assuming its negative working capital unwinds. CEO Jozsef Varadi goes even further:
“If we wouldn’t fly a single flight, we could still finance ourselves for about three years. Many of our competitors have only one to two months of liquidity, many are already trying to get loans or are negotiating with governments for financial aid, otherwise they will not be able to survive. And even if they get that help, their plans will change dramatically because the money will have to be repaid. We expect that weaker airlines will disappear or continue to weaken, while stronger ones will be able to run into significantly better market positions after the crisis”
Even before Corona hit, legacy carriers in Europe, with their broken balance sheets and larded cost structures, were going bankrupt left and right, while Wizz and Ryanair, with unit costs ~1/3 those of the incumbents, opportunistically seized on the distress to open bases and pick up routes. Post-Corona, I think Wizz goes back to generating high-20s returns on invested capital and growing its passenger count by high-teens (vs. mid/high-single digits for Ryanair) as it serves a region of the world, Central and Eastern Europe, where the proportion of the population who has ever flown is 1/3 what it is in Western Europe. If I’m right, then the stock, which currently trades at 8x pre-Corona maintenance free cash flow, is silly cheap.
Credit Acceptance Corp (3%): in the subprime auto industry, lenders typically underwrite for volume, levering up high-single digit returns on capital to generate mid-teens before then blowing themselves up when conditions turn. Credit Acceptance works the other way, targeting a minimum low double-digit return hurdle and letting volumes fall where they may. The company’s incentive structure, which compensates the CEO with shares that vest over 15 years based on cumulative improvement in economic profits, encourages value-accretive underwriting.
In good times and bad, management’s initial estimates of collection rates (the percent of face value that it expects to collect) have been remarkably accurate. During the Great Recession collection rates on prior year loan vintages came right in line with its starting estimates. And while its wounded peers were forced to contract their lending activities in 2009/2010, Credit Acceptance expanded its loan book, taking advantage of constrained supply conditions to later realize outsized returns. The company has been profitable every year over the last 15 years, averaging 14% returns on capital and in no year generating a return less than 11%. Since 2011, a period of intensely competitive underwriting, the company has earned 12% incremental returns on deployed capital.
The company’s recourse debt is well laddered, with $659mn of pre-Corona net profits offering a substantial cushion relative to the $152mn of maturities coming due this year (with the next maturity of $250mn not coming due until 2023). Its 2019 operating profits of $1bn cover $200mn of interest expense (mostly paid to holders of non-recourse securitized debt) many times over.
Booking Holdings (5%): Half of Booking’s cost structure is comprised of variable performance marketing spend (Google ads) that can be rapidly pulled back in the event of a dramatic demand reduction. Revenue could be cut in half and Booking would still be EBITDA profitable. Meanwhile, the company’s $8bn of gross cash at year-end easily absorbs the $1bn debt maturity coming due in June and $2bn of negative working capital.
One structural concern might be that independent hoteliers in Europe, already fed up with the commissions they pay Booking, have been so badly wounded by COVID that they will be consider other distribution channels. What often gets lost in this argument (which seems to recur time and again) is that Booking provides incredibly valuable services to small hotel operators, who don’t have the capital or expertise to dynamically price their rooms or orchestrate large scale ad campaigns on search engines and social media. Booking acts as a centralized entity that can scale the marketing and back-end costs that hotel operators would otherwise have to individually bear. Of course, nobody wants to pay commissions; nobody wants to be aggregated. The point is that there are few viable alternatives for a hotelier with high fixed costs operating in a fragmented industry selling perishable, low marginal cost inventory.
Well except maybe Google, whose growing presence in travel has rightly raised disintermediation concerns for the OTA industry. I worry about Google too. But over half of Booking’s traffic is direct and its direct traffic is growing faster than paid traffic. And while Google is crowding out organic search results with paid advertising, only ~5%-10% of Booking’s revenue comes from the former. That the vast majority of Booking’s leads from Google are paid is an important distinction because travel is one of the latter’s largest ad categories and within travel, Booking is the largest advertiser, which makes it an important customer rather than a free rider on the company’s search infrastructure. Booking and Google enjoy a profitable symbiotic relationship, one that I don’t think the latter will be eager to compromise.
As always, I stand ready and willing to change my mind based on new information or re-assessment of old information. But for now, I think 13x pre-Corona profits is far too cheap for such a dominant, well-moated cash generative business.
HEICO A Shares (6%): The severe strain that airlines are now experiencing may accelerate demand for HEICO’s PMA parts, which offer equivalent safety and efficacy as the original for 50%-70% of the cost. Despite the low IP content of its products, HEICO is a better business than one might think. More so than the FAA approval process, the time it takes for the purchasing and engineering departments of an airline to sign off on the purchase of a new PMA part constrains the number of new products a PMA manufacturer can launch in each year. Even with the relationships it has nurtured with airlines over the decades, HEICO is limited to ~600 to ~700 annual certifications. It would take some time for a competitor, even a well-resourced one with HEICO’s impeccable track record, to recreate the 11k parts that HEICO has qualified in its portfolio. And once an airline has qualified one of HEICO’s PMA parts and worked it into its rotation, there is no reason for it to revert to the original. This is true even if the OEM retaliates by price matching because the airline prefers having a secondary supplier in the market as a counterbalance.
With just $570mn of gross debt on its balance sheet against its $522mn of EBITDA (27% margins) and no meaningful maturities coming due until 2023, I don’t think HEICO will have any issues surviving what is sure to be a violent near-term contraction in air travel and maintenance spend, after which time the company will continue reinvesting its cash flows in acquisitions at reasonable prices and churning out high-teens returns on capital.
You’ll notice that I sized Wizz and Credit Acceptance, whose survival prospects I consider dicier than those of Booking and HEICO, at just ~3%-4%, smaller than our typical core position size of ~7%-8%. Of course, if I didn’t think those companies could survive a draconian downside scenario, we wouldn’t own them.
Still, we’re living in crazy times, you know?…the left tail may be much fatter than I believe. Maybe Wizz’s fleet is grounded for years. Maybe a huge swath of CACC’s dealer base goes bankrupt at the same time that investors stop funding subprime auto securitizations and the company’s borrowers, financially troubled consumers most vulnerable to Corona’s economic impacts, completely halt their car payments, giving rise to cataclysmic revenue declines and deep credit losses. I don’t think either of these scenarios are likely, but every now and then I think about how, at least according to some accounts, one day a virus made its way from a bat to a pangolin to a human in a seafood market 6,000 miles away, and about how now 4 months later, global supply chains are crippled, trillions in wealth have been destroyed, and 10s of millions will be out of work, and I am stopped dead in my tracks, awed by the scope of my ignorance about how bad things can get. And so, in the event I’m wrong about their ability to make to through, Credit Acceptance and Wizz are each sized to avoid calamitous impairment to our portfolio. Moreover, it is possible, likely even, that I will adjust position sizes (up or down) in the coming months as I think more about how COVID impacts the long-term business prospects of the companies we own.
Ok, what else. Back in January, I consolidated what I considered somewhat redundant exposures, selling Ryanair and adding to Wizz, selling Visa and adding to Mastercard. I also sold our entire stake in MercadoLibre on valuation concerns. In March, I sold our stake in Interxion (INXN), whose acquisition by Digital Realty was announced last October, and reduced our positions in Amazon (AMZN) and Everbridge (EVBG), both fine companies that carried too high an opportunity cost to hold given other opportunities that presented themselves. I added to Wix, whose enterprise value was at one point was 2/3 covered by the PV of cash flows from its existing subscriber base, giving the company very little credit for future growth; and The Trade Desk, whose near-term earnings prospects will be dented by reductions in ad spend but whose long-term positioning as a major buy-side platform in programmatic advertising remains as promising as ever. During the quarter, I also added to our holdings in Guidewire (GWRE), Microsoft (MSFT), and Mastercard (MA).
And then there’s Charles Schwab, whose stock was justifiably hammered as rates careened toward 0 last quarter. I added to our position during the quarter. When taking into account steady mid/high-single digit asset growth; the reallocation to cash that Schwab typically experiences during market downturns; significant cost synergies from the TD Ameritrade deal that I discussed last quarter, I can impute acceptable low-teens returns from the current price, and if rates bounce off today’s depressed levels, well, then things get really interesting.
I sold our entire position in Protector Forsikring in late January as a key pillar of my thesis – that the insurer could push through double-digit rate hikes without sacrificing volume and it instead announced huge churn in a major renewal period – seemed impaired. In my Protector post from January 2019, I wrote:
“When I come across an underwriter, like Protector Fosikring, that is profitably growing premiums by 20%/year when its peer group is growing 1-4%/year, the first flag I raise is colored red, not green. More likely than not, it is underpricing risk or accepting risk that its competitors are wise enough to avoid”.
I was willing to make an exception in Protector’s case because the reserve issues that caused its stock to sell off seemed one-off and were followed by widespread insider buying. But underwriting losses have continued to mount and it now seems clear that the company will need to raise prices across the board to restore profitability. This might be fine if customers acquiesced, but elevated churn now casts this assumption into doubt.
There is a final thing I’d like to discuss. I have always treated cash as a residual rather than as a market call, so the fact that ~30% of the fund has, on average, been sitting in cash since the fund’s inception reflects my failure to find attractive things to buy. Having 19% of the fund in cash going into March happened to work in our favor this quarter, but to use this as evidence that holding cash will help our returns over the long-haul is an example of drawing the wrong lessons from a favorable near-term outcome. That I have held so much cash since inception is a process error. Assuming I have a talent for stock picking (time will tell), I believe we will generate better (albeit more volatile) returns over the next 20 years by being more fully invested and actively rebalancing.
Thanks for all your support. Be well; stay safe,
Please visit the blog for more detail on stocks mentioned in this letter.
 A PMA part is an aircraft or engine component deemed by the FAA to be of comparable quality to the one produced by the original manufacturer
 from 2007 to 2009, Schwab’s net interest yield was cut in half from 4% to 2%, but net interest revenue declined by “just” 27% because of offsetting growth in client cash balances.
This report is being furnished by Compound Insight LLC (“CI”) on a confidential basis to existing limited partners in Forage Capital, LP (the “Fund”) and does not constitute an offer, solicitation or recommendation to sell or an offer to buy any securities, investment products or investment advisory services. This report is being provided to existing limited partners for informational purposes only, and may not be disseminated, communicated or otherwise disclosed by the recipient to any third party without the prior written consent of CI. Any offer or solicitation of an investment in the Fund may be made only by delivery of the Fund’s confidential private offering memorandum to qualified investors.
An investment in the Fund involves a significant degree of risk, and there can be no assurance that its investment objectives will be achieved or that its investments will be profitable. Unless otherwise noted, the performance results of the Fund included in this report are presented on a net-of-fees basis and reflect the deduction of, among other things, underlying management and performance fees and expenses as well as brokerage and/or custodial fees and expenses. Performance results also include the reinvestment of dividends and other earnings. Certain of the performance information presented in this report are unaudited estimates based upon the information available to CI as of the date hereof, and are subject to subsequent revision as a result of the Fund’s audit. An investor’s actual performance and actual fees may differ from the performance information shown due to, among other factors, the timing of capital contributions and withdrawals. Past performance is not necessarily an indication of future performance or profitability. An investment in the Fund is subject to a wide variety of risks and considerations as detailed in the confidential private offering memorandum of the Fund.
References to S&P 500 Total Return Index are for informational and general comparative purposes only. There are significant differences between this index and the investment program of the Fund. The Fund does not invest in all or necessarily any significant portion of the securities, industries, or strategies represented by this index. References to this index do not suggest that the Fund will, or is likely to achieve returns, volatility, or other results similar to it.
This report and the accompanying discussion include forward-looking statements. Forward-looking statements are not a promise or guaranty about future events. All statements that are not historical facts are forward-looking statements, including any statements that relate to future market conditions, results, operations, strategies, or other future conditions or developments and any statements regarding objectives, opportunities, positioning, or prospects. Forward-looking statements are necessarily based upon speculation, expectations, estimates, and assumptions that are inherently unreliable and subject to significant business, economic, and competitive uncertainties and contingencies.
The information in this presentation is not intended to provide, and should not be relied upon for, accounting, legal, or tax advice or investment recommendations. Each recipient should consult its own tax, legal, accounting, financial, or other advisors about the issues discussed herein.