Terry Smith on Mistakes & Learning

Sir John Templeton, who founded the eponymous investment management group, once said, ‘The four most dangerous words in investing are: This time it’s different’. He was pointing out that there are always people who are willing to rationalise outbursts of investment mania but they are always proven wrong when the bubble bursts and investment fundamentals reassert themselves. We have seen this before, not only in the Dotcom boom and bust, but in other examples such as the Japanese market in the late 1980s. Then we were told that the PE of over 50 on the Nikkei Index was OK because Japanese accounting was conservative. In fact the market was just over-valued. After the subsequent fall in the Nikkei it took until 2024 for the index to regain the peak it attained in 1989. When companies and/or investors are encouraged by soaring share prices and valuations to believe that capital is almost free, some disastrous investment decisions follow. They seem to act as though the cost of the capital that companies are investing is to some degree the reciprocal of their PE ratio. So, a PE of 50 equates to a cost of capital of 2% (100÷50). This is utter nonsense. The cost of equity does not vary inversely with the valuation and is perhaps best estimated by the cost of so-called risk-free capital, being the yield on long-dated government bonds plus what is called an equity risk premium.

We sold McCormick as we had been disappointed by the slow response which the company exhibited in its ability to pass on input cost inflation so compressing its margins, together with its exposure to own label competition which has stiffened as inflation has caused consumers to trade down. We began purchasing Apple two years ago at about $156 a share when its P/E was below the S&P 500 average and the growth in service revenues had somewhat convinced us that the much talked about ecosystem, tying its users to the products, might really exist. We correctly foresaw a number of reporting periods ahead when sales growth would be lacklustre and so bought a small stake hoping to add to it as the poor sales performance came to pass. We were right about the sales performance — its sales grew just 2% last year — but wrong about the share price which rose strongly, placing the shares on a rating about 50% higher than the S&P 500. We were not going to buy more stock against that background and it was occupying a place in our portfolio and so we sold our stake. We started purchasing stakes in Atlas Copco and Texas Instruments during the year.

Quantitative Easing (‘QE’), so-called ‘printing money’ in which central banks created money to purchase assets, starting with government debt but eventually ranging into corporate debt and equities. As an aside, quite how it aided the economy of either Japan or Switzerland for their central banks to buy international equities is beyond my grasp. This was combined with low, no (Zero Interest Rate Policy — ZIRP) or even negative interest rates (NIRP). These measures I have collectively christened with the generic term ‘easy money’. Attempts to suppress volatility will only exacerbate it in the long term. If you count the current events, we have now had three economic and financial crises this century and it is still in its first quarter. This would seem to illustrate that attempts to expunge volatility from the financial system are actually producing the opposite of the desired effect. They breach the rule for what you should do if you find yourself in a hole. This is hardly surprising given that the central banks were aiming at the wrong targets. Central banks were attempting to maintain a benign level of consumer price inflation but ignored asset price inflation caused by their actions. Some also adopted employment targets that were not or should not be part of their remit. One of the problems of easy money is that it leads to bad capital allocation or investment decisions which are exposed as the tide goes out.

triple A credits when they were really triple Z. You can’t improve the quality or liquidity of an asset by putting it into a structure. The other problem with the policy of easy money was that it had to end eventually, but not before it had one last hurrah. There were half-hearted attempts to reverse QE in particular by lowering central banks’ bond purchases but when the stock market unsurprisingly reacted badly in the so-called ‘taper tantrum’ in 2013, these were abandoned. Then in 2020 came the pandemic and central banks reacted to this by enacting that good old saying ‘To a man with a hammer, everything looks like a nail’. They decided that they should double down with their new toy, QE, which seemed to work so well in the Credit Crisis without any nasty side effects, well none that had yet become apparent, and apply an almighty stimulus. This was applied when there was no problem with demand or the banking system. It was just that people were locked up in their homes and unable to spend on bricks & mortar shopping, travel and entertainment and the global supply chain was malfunctioning, leaving consumers with pent-up savings waiting to be spent. What happened next may be an example of Sod’s Corollary to Murphy’s Law: • Murphy’s Law: What can go wrong will go wrong. • Sod’s Corollary: Murphy was an optimist.

One other topic which I want to cover this year is share-based compensation and especially its removal from non-GAAP (Generally Accepted Accounting Principles) profit figures. Share-based compensation has become an increasingly prominent part of some companies’ expenses in recent years, especially among companies in the Technology sector. If we take for example the 75 companies in the S&P Dow Jones Technology Select Sector Index, share-based compensation expense expressed as a percentage of revenue has gone from an average of 2.2% in 2011 to 4.1% in 2021. This may not seem like much of an increase, but keep in mind that during this period revenue for this set of companies had almost quintupled on average. There is nothing wrong per se with compensating employees with shares. In fact, there is a legitimate reason for doing so: it may help to align the interests of employees with those of shareholders. I want to focus on how share-based compensation is accounted for or, more accurately, how it is not accounted for in companies’ non-GAAP earnings figures.

Researchers found that the rate of ovarian cancer was not significantly different between those who did and did not use talcum powder. The study concluded that the rate of ovarian cancer amongst those who used talcum powder was 61 cases per 100,000 people per year, compared to 55 cases for those who have never used it. Therefore, the current scientific evidence would imply that J&J did not falsely advertise the safety of its Baby Powder, as there is no evidence of asbestos in its talcum powder and talcum powder itself has not been found to cause cancer. An important lesson from this example is that negative impacts are never clear-cut and the devil is in the detail. This is especially true when assessing the extent of a company’s responsibility for a negative impact. The headlines can sometimes give the wrong impression of a company’s guilt or exaggerate the degree of control a company has. This is why we don’t automatically exclude any company that has a RepRisk Indicator score above a certain level and why any assessment of a company with a high RRI needs to look at the details. This example also raises the question that if talcum powder did cause cancer, as investors, what should our stance be toward corporate responsibility in the face of questionable scientific evidence? Clearly, corporations should follow regulations. They should do due diligence in ensuring their products are safe and effective.

nothing is risk free. At what point are risks so low that they are negligible? What is the responsibility of corporations to disclose possible, but not proven, tiny risks? If they do, this can lead to “alarm fatigue,” where consumers learn to ignore warnings because they are everywhere. These questions become especially relevant to medical device makers or personal care products, which don’t go through as much (or any) of the rigorous testing that the FDA requires for drug manufacturers. How much testing should or can be done on a product before it is released and at what point has a company done all it can to identify and assess these risks before they are no longer held responsible is an open question. RepRisk also doesn’t look at any positive impacts, which are particularly relevant for a company like J&J which has the large positive impacts that are too often ignored. We believe that when assessing the impacts a company has it should be done on a net basis, as a company will get a lot of publicity when things go wrong but significantly less for the good things it does every day. In 2018, J&J provided 39,000 people with access to tuberculosis treatment and 52,000 people access to HIV treatment, trained 105,000 health workers in 67 countries and invested $11bn in R&D to develop new treatments that help patients live better and longer lives.

The reality is that the only type of fund which can guarantee 100% liquidity on demand is a cash fund, and I presume that is not what you wish us to invest in. But I suspect you will find it hard to find more liquid equity funds than ours. It tells you much about its liquidity that some of the least liquid stocks we hold are the FTSE 100 companies, Intertek and Sage. Another question which arises from the Woodford incident is the question mark over so-called ‘star’ fund managers, a label the press seems obsessed by. I can’t say I like the term, it strikes me as equally inappropriate as the term ‘beauty parade’ which is used when selecting professional advisers, many of whom do not seem to me to have obvious photogenic qualities. I think this concern is focused on the wrong issue. I think it makes no more sense to avoid funds run by ‘star’ fund managers any more than it does to avoid supporting sporting teams because they have star players. The trouble arises not because teams have star players but if the star tries to play a different game to the one which delivered their stellar performance. Would Juventus do as well if Cristiano Ronaldo played as goalkeeper? How is Usain Bolt’s second career as a soccer player going? Neil Woodford made his name as a fund manager at Invesco Perpetual with his successful Income Fund. In the course of this he took two high profile negative positions on sectors.

In the run up to the dotcom bust in 2000 he seems to have seen what was coming and avoided investments in technology, media and telecommunications stocks which was a major success. He also paired this with taking positions in some of the old economy neglected stocks which had become de-rated during the dotcom mania. Similarly, in the run up to the Credit Crisis he decided not to hold bank stocks. However, when he opened his own fund management business he took positions in a wide range of companies — AA, AstraZeneca, Capita, Imperial Brands, Provident Financial and Stobart are some examples. There is no common theme that I can detect to those companies, other than the fact that they all subsequently fared badly. This was supplemented by a raft of unquoted investments in start-ups and biotech. My suggestion is that what went wrong is that Neil Woodford changed his investment strategy. In the technical jargon of the industry, he engaged in ‘style drift’. The problem wasn’t that he was regarded as a star but that he changed his game. This style drift actually started when he was still at Invesco Perpetual in that his Income Fund began to accumulate large stakes in small illiquid companies and unquoteds, but this was taken further once he had his own firm.

You or they might well counter by saying that this past outperformance is all very well but it does not help you in making a decision on whether to own our Fund from today, which must surely be determined by its future performance or as the legalese goes ‘Past performance is not necessarily a guide to future performance’. I think the key word in that sentence is ‘necessarily’. Let me offer a couple of thoughts on that. The first problem is of course that the commentators upon whom you might rely may simply be wrong.that:

• The UK would vote for ‘Remain’ in the Brexit referendum • The UK would enter a recession immediately if it voted to ‘Leave’ the EU • Donald Trump would not become President • Narendra Modi would not become Prime Minister of India • Narendra Modi’s economic reforms would fail • Theresa May would have such a resounding victory in the 2017 election that Labour would disintegrate • Angela Merkel would sweep to victory in the German elections • President Trump’s tax reform bill would not be passed by the US legislature In some cases, they have a ‘Full House’ having made all these predictions. The fact that they have been shown to be comprehensively wrong does not seem to stop them from giving us the dubious benefit of further predictions. In this regard they remind me of the broker who was always wrong and who is mentioned in the book ‘Hedgehogging’ by Barton Biggs, the strategist and hedge fund manager. Biggs found him useful to talk to because once the broker had given his views on what would happen or what to do, Biggs knew that the opposite was bound to be correct. For what it’s worth, my diagnosis of the problem for these commentators who seem to emulate this broker is that they are experiencing role confusion. They seem to have forgotten that their role is to report events accurately and have decided that instead they need to influence the outcome to one they desire.

They also seem to have missed the point that voicing your views in an echo chamber is not likely to lead to a challenging debate in which to test your opinions. Thankfully, I spend little or no time trying to apply predictions about macro events in order to manage our portfolio. However, that does not mean that I do not think about them. As I have maintained for most of the decade since the Financial Crisis, looking back to the Great Depression for an analogy that would enable us to understand these events and form a view of how they may unfold is probably a mistake. A better analogy may be the Long Depression of 1873–96 when a new industrial power came on stream and caused a wave of deflation as it could manufacture goods cheaper than in the Old World. That industrial power was America after the Civil War. The Long Depression was also preceded by a collapse of part of the banking system. Sound familiar? The wave of deflation we have been experiencing has been caused by a number of factors. These include the rise of China as the world’s greatest industrial power, other cheap manufacturers (South Korea, Thailand, Vietnam, India and Malaysia for example) and the offshoring of manufacturing to cheap manufacturers under free trade agreements, such as Mexico under NAFTA, which so exorcises President Trump.Depression

change you expect in market conditions. The problem is this; if you choose the one hundred and five 400-metre runner route I presume that to make the contest against the marathon runner realistic you have to carry a baton that you hand over to the next runner. This is the equivalent of you making the decision to sell all your high quality stocks and switch into somewhat cheaper (although maybe not cheap) cyclicals and value stocks. However, I seem to recall that very often that baton gets dropped, or the changeover is not made within the allowed zone and the team is disqualified. I suppose the investment version of this is that you get the timing of your switch wrong or you sell one strategy but remain in cash. The problem in trying to apply this sprint strategy in the real world of investment is even worse. In a relay race the runners for each stage are selected in advance. Whereas in an attempt to apply this technique in investment you would need to select whom you wish to receive the baton as you enter the changeover area each time. After all do you know in advance whether you want to go from high quality consumer staples to financials, commodity stocks or industrials, emerging markets, bonds or some combination of these? The scope for fumbled handovers is endless. And you have to do it many times to succeed with this approach. Moving on to review the outcome for 2017 in terms of our Fund’s strategy.

We have often been told that if a company has two divisions and one is in a slow growing segment and one is faster growing (like PepsiCo with soft drinks and snacks) then if the two are separated (as Nelson Peltz suggested to PepsiCo) the faster growing one will attain a higher stock market rating once on its own. This is probably true, but won’t that be compensated by a lower rating on the slower growth division? Of course not for the activist who intends to sell out as soon as possible. Thankfully in our view, on this occasion Mr. Peltz was unsuccessful and PepsiCo remains a drinks and snack business, which is not to say that we think everything is fine with PepsiCo’s management or that Mr. Peltz is always wrong, of which more later. Leveraging up the balance sheet to buy back stock is a frequent demand of activists and is invariably described as ‘returning cash to shareholders’ and not only when it is suggested by activists. The correct description for this action should be ‘returning cash to exiting shareholders’ as we remaining shareholders don’t receive any of it and this perhaps best encapsulates the problem we identify with this practice. Those of us who actually seek to own the company and remain shareholders see debt raised to take out shareholders who wish to exit. It is beyond us why we would want that to happen unless the shares purchased are demonstrably cheap.

scope for subsequent increases and the lack of effect on long term rates. In which case worries about the effect on so-called bond proxies may prove to be overdone. Secondly, what would these commentators have you do about this possible adverse impact on so-called ‘bond proxies’? Presumably they recommend selling them in view of this predicted disaster and investing your money elsewhere. Leaving aside the commentator who suggested that the answer was to invest in a fund which is ‘more immune to future market performance’ (seems like an overly modest target - why not just find one that only ever goes up?), the most common suggestion, it seems, is that you should consider switching into more cyclical stocks because they are more lowly rated and their returns are too volatile to be considered as bond proxies. Switching into cyclical stocks in anticipation of a rise in interest rates, what could possibly go wrong? As ever, spotting potential problems with our or any other investment strategy is not that difficult. In all my years in business I have never found that identifying a problem is quite as difficult as solving it. Likewise, suggesting what it is you should switch into that is immune from problems which may result from an interest rate rise is a bit more difficult. However, it seems likely that sooner or later the ‘stopped clock’ commentators will prove to be right and our Fund will experience a period of underperformance. What to do about that?

been postponed. We took this as a bad sign in a banking market which is exemplified by a cartoon which shows a man sitting in front of a bank manager (you can tell this because there’s a sign on the desk saying “Bank Manager”) who says “I’d like to borrow some money” to which the Bank Manager replies “What a coincidence, so would we.” There is clearly nothing wrong with Domino’s but plenty wrong with the banking industry on which it was reliant for its refinancing. In the event, Domino’s proved us comprehensively wrong. Not only did it manage to refinance but they did so on terms which enabled it to pay a $3 per share special dividend. So I did what you should always do, but we so rarely manage to do, when we get it wrong a) admit this - most importantly to yourself; and b) reverse the decision. So Domino’s was repurchased Fortunately there was a period of share price weakness after the refinancing which enabled us to do this on reasonable terms but frankly that does not matter as much as whether the shares were still good value when we repurchased them, which we believe they were. It is always a mistaken strategy to wait for the shares to get below the point at which you sold them before repurchasing, or the even more common trait of waiting for a loss-making share purchase to get back to break even before selling.

Over the long run, it has contributed a higher percentage of equity performance than share price appreciation. But I would caution against a blind search for higher yields. The current record low interest rates and bond yields have produced a desperate search by investors for yield. The investment industry stands ready to supply products to satisfy any craving by investors, not always to their advantage. Investment flows have started to gravitate to higher risk bonds such as junk bonds and emerging market debt as government bond yields in the supposed safe haven countries have shrunk towards zero. The yield on US high yield or junk bonds sank to 6% at the beginning of 2013, the lowest ever recorded. New issuance has boomed in high yielding real estate investment trusts, and so-called master limited partnerships in energy stocks and pipeline companies (I wonder how many investors can explain how they work). Even Collateralised Loan Obligations (“CLO”s), part of the toxic alphabet soup of instruments which helped start the Credit Crisis have been making a comeback with issuance trebling in 2012. How soon we forget. Equity investors are far from immune from this trend. For many investors the search for yield is satisfied by investing in an income fund which invests in high yielding equities. This can be a mistake.

There is nothing in the performance of Domino’s which causes us the slightest concern about this but there is plenty wrong with the banking system which will be required to provide the refinancing. As a result we hope to have the opportunity to become investors in Domino’s again. The net result this was that the Total Expense Ratio of the Fund was 1.2%. We hope to reduce that in future. The historic dividend yield on the Fund at year end was 2.4%. This dividend was covered 2.6 times by earnings. There is only one stock in the Fund that does not currently pay a dividend. This is significant: it is becoming clear that dividends are likely to provide a more significant portion of the total return on equities in the future than they did in the equity bull markets of 1982-2000 and 2003-07. The current yield on the Fund may not fully reflect its dividend paying capabilities as some of the companies also utilise share buybacks. During the course of the year we published some research on share buybacks (“Share Buybacks-Friend or Foe?” April 2011-available on the Fundsmith website) in which we concluded that buybacks were rarely accompanied by any reasoned justification; that they had become almost universally regarded as a good thing and contributing to shareholder value irrespective of the price paid or the valuation implied, which simply cannot be true; and in many cases their timing was poor.

During the year we wrote to the management of those companies within our portfolio which have engaged in share buybacks to ask for some insight into their rationale. The responses ranged from prompt, personalized (by the CEO) and well reasoned to being completely ignored. We regard the greatest risk for our investors after the obvious potential for us to buy the wrong shares or pay too much for shares in the right companies, as being reinvestment risk: we seek to buy companies which deliver high returns on capital in cash. What the management then does with these cash returns is one of the major factors affecting future returns on the portfolio. Management faces three main options for deploying these cash returns: return cash to shareholders, invest to grow the business organically or make acquisitions.for

What happened next even surprised me and I thought I had lost the capacity for such an emotion in the face of the shenanigans of the financial services industry. Practitioners within the ETF sector reacted with a fury which can only be generated by two factors: 1) the criticism was accurate and/or hit a nerve; and 2) it was in danger of derailing a large gravy train. Some ETF practitioners suggested that I was criticizing ETFs because of concerns about the impact the growth of ETFs would have on the active fund management sector in general and Fundsmith in particular. This response is not just wrong it is preposterous for two reasons: 1) Fundsmith’s market share of the active fund management sector is so small that I do not possess a calculator capable of getting enough zeroes to the right of the decimal point to calculate it.could

Mr Forename Surname Company Name Address line one Address line two Address line three Address line four Monday 10th January 2011 Dear Fellow Investor, This is the first annual letter to owners of The Fundsmith Equity Fund. Fundsmith opened for business on 1st November 2010, and we are critical of attempts to measure investment performance over short periods of time. Two months is not a short period, it is a ludicrously short period to do so. However, I thought that this letter is a good opportunity to give you a flavour of the reporting which is likely to occur in years to come. From 1st November to 31st December 2010, The Fundsmith Equity Fund rose by 6.14% net of fees. This compares with some common benchmarks as follows: Fundsmith Equity Fund 6.14% MSCI 7.99% MSCI EAFE 5.76% FTSE100 4.40% Long Bond (10 year UK Treasury) -2.57% Benchmarks are useful for measuring performance, provided a long enough time scale is used. Problems arise when fund managers start to use them for portfolio construction. At Fundsmith we do not endeavour to track any index or to minimise our “tracking error” versus any index (even the use of the expression tracking “error” tells you that an active fund manager has the wrong mindset). The Fund underperformed the MSCI and outperformed the MSCI EAFE-the difference being in the performance of US stocks which are included in the former but not the latter. It outperformed the FTSE100 and long bonds.