At Graham & Dodd's Annual Breakfast in 2022, Michael Mauboussin led a Q&A session with GEICO CEO and Warren Buffett's investing partner Todd Combs. Todd Combs provided his insight on defining a “good business,” how he thinks about the longevity of businesses, the accuracy of financial statements, and a unit cost centered valuation approach. The interview is filled with incredible answers including the three questions Warren and Todd used to find and make their investment in Apple in Q1 2016.

Interview Transcript

Todd Combs opened his discussion on investing by equating it to Einstein’s theory of intelligence, which he said ranges from smart, intelligent, brilliant, genius to simple. Combs states that while investing is simple, it is not easy.

During one of Todd Combs and Warren Buffett’s famed Saturday afternoon living room chats, the two posed the following question as a means of valuation: if you take a business, what is your level of confidence in predicting what it looks like in five years?

Buying a whole business requires varied degrees of scrutiny into factors which cannot be evaluated using financial modeling formulas, such as: level of capital required, management style and efficacy.

Combs recalled the first question Charlie Munger ever asked him was what percentage of S&P 500 businesses would be a “better business” in five years. Combs believed that it was less than 5% of S&P businesses, whereas Munger stated that it was less than 2%. You can have a great business, but it doesn’t mean it will be better in five years. The rate of change in the world is significant, which makes this exercise difficult, but this is something that Charlie, Warren and Todd think about. When Combs started at Berkshire, they had a 7/10 confidence on the businesses outlook for the next five years. The nature of the world is that things are constantly changing, and Todd says they are right on maybe 1/10 predictions.

Michael asked if there are traditional measures that Combs/Berkshire look at to indicate good business performance? And how does Combs/Berkshire assess that quantitatively? Combs explained how one question is constantly asked, usually daily, and that is if the moat is wider or narrower on any of their businesses.

98% of what Buffett and Combs discuss is qualitative. If something is 30x earnings you can calculate what it will have to do to get to run rate earnings. The worst business grows and needs infinite capital with declining returns. The best business grows exponentially with no capital.

At Berkshire there’s a focus on owner’s earnings. It’s a good sign when the owner’s earnings are close to the reported earnings. Owner’s earnings can be calculated in the following way: (a) reported earnings plus (b) depreciation, depletion, amortization and certain other noncash charges less (c) the average annual amount of capitalized expenditures for plant and equipment, etc., that the business requires to fully maintain its long-term competitive position and its unit volume. (If the business requires additional working capital to maintain its competitive position and unit volume, the incremental spending should also be included in section (c). However, businesses following the LIFO inventory method usually do not require additional working capital if unit volume does not change. LIFO would give the appearance of an additional need for working capital because inventory is held at a cost less than average, and, or FIFO. Buffet is saying not to confuse that with a working capital need.)

Combs goes to Buffett’s house on many Saturdays to talk, and here’s a litmus test they frequently use. Warren asks “How many names in the S&P are going to be 15x earnings in the next 12 months? How many are going to earn more in five years (using a 90% confidence interval), and how many will compound at 7% (using a 50% confidence interval)?” In this exercise, you are solving for cyclicality, compounding, and initial price. Combs said that this rubric was used to find Apple, since at the time the same 3-5 names kept coming up.

Businesses are run by people, and Buffett says he likes taking the cash flow and removing it from managers and investing it himself. There’s a known adage, when looking to buy a business: look to buy a business a dummy can run, because eventually a dummy will run it.

The discussion then shifted to how Combs thinks about understanding if a company is more focused on internal growth vs external growth. Combs mentioned that business destruction has accelerated and that every business has become a data and technology company and to stay aware of what parts of the business are dying.

Charlie famously says you get what you incentivize. If a management is incentivized to appeal to Wall Street, it might behave in a value destructive way.  Combs said that it’s a red flag if a company is too focused on external relations. Combs mentioned how his responsibilities are great, and his time is scarce, but if he was 150% dedicated to analyzing companies, he’d focus on delta reports, and see what management is changing from year to year. In 2005, Bear Stearns changed something very essential in their reports while saying something different to the street.

A big signaling effect for Combs is when management changes the key performance indicators for which it will get compensated by, presumably because management won’t get compensated if the KPIs are left as is.

Combs estimates that in the fortune 500, approximately 20% of companies are changing the incentive metrics for management. That’s not an accident.

Every time Combs meets with a company, there are two questions he always asks management: (1) How long do you spend talking to investors, and (2) what would you be doing if you were not publicly traded? The median response is 25% of the time is spent talking to investors. In response to the second question management usually lists a number of things that make a lot of sense, and Combs then proceeds to ask why they don't do that, and they say because they feel handcuffed. When management is focused on the quarterly performance, and they don’t have the proper time horizon, they are not empowered to do the right thing. As a fiduciary you are setting yourself up for failure if you don’t have the right time horizon.

Categorically, founder led businesses outperform because the founders are the fiduciary, and they have a longer time horizon.

On Valuation

How does Combs look at valuations? What significance does valuation have in his investments, and how does it help him screen for companies?

Combs began by describing a three-legged stool and explaining how you can't ignore any one leg (in this case valuation). You have to pay attention to valuation. You have to have a framework where you are getting a financial return. Sometimes earnings are misleading. David Einhorn once wrote about this, that there's an arbitrage in situations when most people are focused solely on financial metrics, you need to adjust for accounting which can be extremely misleading. For example, at Berkshire things are marked to market but not considering real earnings. You can have SAAS businesses incurring 100 cents on the dollar in expenses on day one and going forward earn whatever the LTV of the customer is. You have to have a view where you can understand what the financial image looks like. For example, “in the early days of Walmart and Amazon, I wouldn’t focus on the fact that the reserves seemed wrong and quite large. Look at where it took them. I build out unit economics (in the case of Walmart look at the individual store), and that is your north star. You backtrack, and you get to CAC, LTV, and so on.”

What is so bad about EBITDA? Combs remarks, “You mean BS earnings?”

Combs then goes on to explain the flaws embedded within EBITDA. Something frequently not understood, not discussed, and consequently not evaluated is maintenance capex. Feynman had this saying “it’s so easy to fool yourself and you’re the easiest person to fool.” CEO’s have a very low grasp of their maintenance capex. Combs further says how most management teams aren't even able to have that discussion because of their confidence in investing for growth. In reality management often is mistaking growth for something that is really maintenance capex. “It might be growth spending, but you’re not growing!”.

A recent Ph.D at Columbia wrote a paper on maintenance CAPEX. The paper found that depreciation understates maintenance CAPEX, he explains it industry by industry. Very few companies talk about maintenance CAPEX. You can have a SAAS company where the maintenance CAPEX is your people and your salesforce, so you can’t ignore it just because you’re not capitalizing those expenses.

When looking at a company Combs begins inside out versus outside in.

Combs explains there’s a lot of outside in. You hear a narrative, you meet with management, and it’s extremely intellectually dishonest. The goal of investing is to have intellectual purity. There’s facts and numbers, and then there is a narrative. Combs tries to wall himself off from the narrative until he has a view. As mentioned before, Combs would start with delta reports. Put together your view without looking at the market cap. Limit the pollution. Then think about the future of the company and pressure test your assumptions.

With regard to quantitative research, Combs doesn’t understand that world as well, if he did Berkshire returns would reflect Renaissance’s returns. Charlie and Todd talk about this a lot. Combs thinks some funds have quant groups, but he’s not so sure about implementing it at Berkshire. He admitted it was somewhat of a cop out answer, but he honestly just doesn't know.

A notable change recently has been a change from tangible to intangible investments at Berkshire. Next, they discuss sticky customers and how that affects valuation.

Combs focuses on fundamental unit economics. “First figure out CAC and LTV for GEICO. You know the data of where they live, the kind of car, and then there is the LTV of the customer. Combs thinks where people go wrong is where they think they know their CAC and LTV but it’s from too high a level. When you slice the data, there are really big pockets where the LTV is negative, and that is where the potential is to improve margins.”

Another example “to shed light on this can be in a large bank, where the overall ROA can be great, but within that you can have pockets of high and low ROAs. So, it’s important to break down and segregate to see actual unit economics.”

With Google, for example, Berkshire missed it despite the auto insurance being the number one and number two groups for paid search. Warren was in a position where he had the right information and the unit economics and yet he still didn’t act on that right information. That’s why he regards it as such a meaningful mistake of inaction.

To some degree, when valuing a business, you disregard the financial statements and accounting, and you focus your attention on the unit economics. Combs even said, “throw the accounting out”! After figuring out unit economics, the first thing Combs does is evaluate P/E, and then understand the owner's earnings. “Look at where you’re at in the lifecycle of the company, and where it is. Then let’s look at where it is going and apply some confidence intervals.”

A traditional value strategy hasn’t done well in the last decade. Why and will it reverse?

Combs likes to think on reflexivity, where there is a feedback loop with low interest rates, great returns from growth. It’s easy to say no to value.

With regard to the few things within the Berkshire portfolio that are growth-ier, Combs says focus on the unit economics.

Mergers and Acquisitions

Most companies struggle to make valuable deals. How do you do a deal that is both strategic and good value?

Charlie once asked Todd about derivatives and Todd said that he really doesn't know much about them. So, Charlie then said, “but if I gave you 1,000, you could probably pick out the best 3.” One does not need to be an expert to see when the scale is large, and it is sometimes better not be an expert to spot a macro trend. “Remove the fish from the fishbowl. If you are only seeing three deals out of ten competitive companies, you’ll probably not allocate capital well. In fact, it might be really bad. 80% of M&A is value destructive. Usually, the price paid is the issue. Strategically, they make sense, but that doesn’t excuse any price. Sometimes 1+1 = 3, the problem is they are paying 5!”

Sometimes companies do deals because other companies do deals. As Charlie says, the best quality is to remain rational.

Combs’ Day-to-Day at Geico and his Managerial Side

Combs has been CEO of GEICO for almost three years. How does he decompartmentalize the investor and management roles?

Combs explains that when it comes to investing, “you don’t need charisma or social skills. You need to have a decent conversation with management. On the other hand, when you are in management, you have to convince people in the company when it comes to initiating change and you have to be good at explaining why change is necessary. As a manager if you are good at communicating you can mobilize people to change the operations, whereas as an investor, you see businesses fail left and right, you have the experience, so theoretically, you should be able to come in and tell people to change. But it doesn’t work like that, and management usually is too charismatic to listen.”

Here are some places where managing and investing overlap:

1. Process vs outcome is similar. Carry around a list of things to get done.
2. Intellectual honesty
3. Facts and numbers vs narratives
4. Everything is a meritocracy as an investor. You need to reinforce that as a CEO.
5. Due diligence, Combs loves “skip level meetings”. That’s where all the information is, speak to the people 3-7 levels down!
6. Get the big stuff right
7. Finding owners who are fiduciaries, (Charlie asked Combs what percent of management are fiduciaries)
8. Transparency and accountability, the people in power are held accountable and are aware of the implications their actions have
9. Avoiding past dependencies, tendency to fall prey to trusting past research and former experience
10. Find areas where you can push a little and change a lot. You want nuclear fusion. Put in 100 units and get out 1,000 or a 1,000,000

How has Combs dealt with incentives at GEICO, has he made changes where he was surprised by the outcomes?

Combs explained how there’s no perfect way to structure incentives. On the one hand, everyone on the team gets a super bowl ring, but you also want to make sure people are compensated by the things they can impact and not the things out of their control. Being private definitely helps. Combs changed the compensation when he arrived. “Compensation can also lead to behavior that is not good, e.g., at Wells Fargo. Anything that is good can be taken too far. GEICO used to have a regional structure, but that was changed. You want a meritocracy that is bespoke and tailored to the individuals, you want to compensate people for alpha, not beta. “

Public vs Private

What does it take to be public and do the right thing?

Combs says that you need to have the north star. Again, as was said earlier, founders usually have this concept best. Lay out the principles. Look at how you do things and why. Are you focused on style, or are you focused on substance? If you are outward focused, you’ll probably focus more on style.

Haven: The Healthcare Joint Venture

Combs was involved in the joint venture with Amazon’s Jeff Bezos and JP Morgan’s Jamie Dimon to fix the healthcare system in the US. What was the mission, why was it dissolved, and what did Combs learn?

Combs remarks that it was a lofty goal. The three partners decided they weren't making the progress they needed to keep it going. They each have a number of programs running internally that are like petri dishes, but the medical system in the US is a very entrenched system. COVID helped the digital disruption, which shows the environment can move much faster than you can control.

Answers to Questions from the Audience

A question was asked how Combs thinks about selling a stock that may be fully valued vs its tax implications, and, on a different note, how he balances his time with being the CEO of GEICO and managing the portfolio?

Combs answered when Warren bought Burlington, he said he should have bought it long before he actually did. His great attribute is knowing when and how to adapt. Like the Keynes quote “the facts have changed, what did you do?” Most of us get more set in our ways. It’s part of being a learning machine to avoid that.

“Generally speaking, we (Berkshire) don’t pay a lot of attention to taxes. Some of it is bespoke. Some are compounders, vs return to par, which is cigar butts which we don’t usually do. Coke is more in the compounder category. “

With respect to his time management, Combs said that he has a couple of jobs. From an allocation perspective, 100-110% at GEICO, 16-hour days there, and he does his investing at nights and weekends, which makes up approximately 25%, and finally bank board meetings and other is 25% (Combs sits on JP Morgan’s board which unlike a traditional company has 8 board meetings a year as opposed to 4).

Bill Ackman asked if Combs could talk about Berkshire’s investment in Nubank. How did the idea come to Berkshire, how was it found, what was the nature of the due diligence, and how do you (Combs) think about its value?

Combs didn’t really want to discuss any names in particular. So he asked Bill if there was something else he would like to ask. To which Bill responded by questioning the moral grounds of investing in a company that makes sugary beverages (Coca Cola or Pepsi) and does a great job showing skinny people playing volleyball on the beach while consuming their product. Then you look at Mexico City, cities around the world, and the penetration of soft drinks, and you get diabetes, obesity, fatty liver disease, etc. This company is effectively the Phillip Morris of this generation. These companies are causing a lot of harm and they are doing a really good job marketing it, you have to believe that at HQ they know what is going on.

Combs wittingly began by saying that the first question looked more appealing now knowing hearing the second one. He also reminded everyone that Coca Cola is obviously not his name. However, Charlie got this question on Wells, and Charlie said something like “we have to invest in the world we live in, and not the world we want.,” “It’s a tough question. We used to have these questions on payday lending. But if you completely cut these things off, then people don’t have access to markets at all.” Combs acknowledged that he didn’t have an answer, and asked if Ackman asked Warren or Charlie?

During times like this, a question about inflation was unavoidable. Someone asked what Comb’s thoughts were on the macro-outlook.

He reiterated that they don’t do macro, and they don’t predict the future at all. What they do instead is take the world they’ve been given in the present, without having to understand the future. They ask if inflation is secular or cyclical. (Warren and Todd were talking for hours this past Saturday and they didn’t talk about inflation for more than 30 seconds.) In a deflationary environment they would ask how this business would perform in 10% inflation. Never just extrapolate the current environment state. Cyclicals always look expensive at the bottom and cheap at the top. Always invert. Combs thinks about it like “a hot water faucet. Sometimes it takes some time to warm up and when you feel it you might gently scald your hand, the more inflation goes the more the Fed is going to be after it, and it can lead to recession. It works like a pendulum.”

Someone asked how Combs/Berkshire views overseas investments? And if the strength of the US Dollar changes things? Additionally, many businesses had to close down due to the Russian invasion. How do they think about geopolitical risks, internally to Berkshire and as an opportunity to buy foreign businesses?

Combs said the biggest issue is size. Many markets have few opportunities, Berkshire will look at them and they’ll have to assess those vs the opportunities it has here. There’s also a currency component when investing in other countries. Combs remarked how Berkshire may have missed some opportunities. With regard to Russia, the biggest overseas operation is Iscar, and in Russia its presence was relatively de-minimis. There might be changes in leadership and things can change, but it's definitely possible to have a large top ten stake in a large overseas company.

Someone asked when Combs joined Berkshire, what were some of the things he looked to accomplish, how that’s going, professionally and personally.

He responded that when he came to Berkshire, he took over Lou Simpson’s portfolio. Over time, after positive performance, he got a couple more billion. In the summer of 2011, the European market imploded. People were worried about Europe disbanding. He was strictly focused on the portfolio. He got a look at every deal that Warren did. In the 200 plus deals that came through to Berkshire every year.

With regard to what Combs wanted to accomplish, Combs said that he was wondering how to add value. He was hired to pick stocks, but Buffett was already pretty good at that. Buffett and Combs had many conversations about the paint industry that led into some things with operating companies. “I was not outcome oriented; I was just process oriented and trying to add value. With GEICO we want to be the #1 insurer, it’s compounded at 15% since Buffett bought it, I want to take it up a level.”

Someone said being in a room of a lot of people who started their careers at Goldman, Morgan Stanley, and JP Morgan what are some of the key takeaways Combs got from the state of Florida securities regulation division.

Combs graduated undergrad from Florida State. He said he didn’t have that option for investment banking, and further said the experience that he got was very robust and probably better than his friends at Arthur Anderson. Combs always wanted to be an investor but started his career working for the Florida State securities regulator. When he worked for the regulator, he worked on a case of insurance fraud. That introduced him to the parallels between insurance and investing. Additionally, at the time his friend was working for Progressive, and he loved it. Combs decided to try working directly in insurance and landed a job at Progressive. Later, Combs interned at a fund when he was at Columbia Business School, and he “was fortunate not to destroy value for people.” After graduating with an MBA, he got a job at a hedge fund and went on to start his own before coming to Berkshire.

The final question asked was about their publicly listed competitor, Progressive, and how Combs/Berkshire competes?

Combs began by saying that he was really into data, and when he was at Progressive, they were doing things that have now become very popular. GEICO has also been very successful in the direct-to-consumer channel. GEICO used to be direct, like consumer banking, Progressive honed their chops in a sector that is more analogous to the commercial banking channel. All top ten insurers were formed prior to WWII. A lot of things have become much more challenging. Combs concluded that “We are investing in data and technology. We have lots of initiatives going on and they take time. We have a good infrastructure and a foundation of great people.”