Going back to the fundamentals of investments and valuations, and to ascertain whether a company's valuation is more than just a quixotic pursuit. Avnish Bajaj (Founder & Managing Director, Matrix Partners India) gives us a piece by piece break down of the valuation techniques and strategies used to analyse and determine a company's real value versus price.
Making Sense of Valuations
Salonie: Hi, and welcome back to Matrix Moments, this is Salonie and I’m here today with Avnish Bajaj, Founder & Managing Director Matrix Partners India. Today’s episode is about making sense valuations and whether it is a quixotic pursuit or not. So how are startups really valued in venture capital Avnish, can you break this down for us?
Avnish: Hi, Salonie. Happy to be back for another Matrix Moment. Another topic that is very near and dear to my heart and I am going to warn the listeners and this is going to be a bit of a combination of some theory, some practice and some views around it, but let’s start by~ let me ask you, why do people invest, why do you invest?
Salonie: To make money?
Avnish: So, what is money? What are you looking for? Well, let me answer that, I mean money is of different forms, I mean people look at capital gains, appreciation in price would be one form of money, and the other would be actual earnings, right? Actually making, getting cash flows and income out of the investment. So, the reality is, as you are thinking about it and let’s chat a bit more on this ~ how do you see investing in a bank fixed deposit different from investing in, let’s say in an early stage company.
Salonie: Higher returns
Avnish: Very good. That’s one way of thinking about it. There is also higher risk. Higher returns are coming because of higher risks but that is exactly the point. The reality is, before we go to valuations, we should talk about why do valuations matter and the reality is that valuations come in only if you are investing and what is the value at which you are investing? So, we say “okay why do people invest?” because they want returns, like you said. Returns vary across asset classes, there are notionally less, not notionally but actually probably in reality, less risky asset classes. You said, bank FD’s, they are less risky. The ultimate so called ‘risk-less’ asset class maybe cash, who knows but cash can actually go down if inflation. So, lot of technical theory comes in. Now, gold- one can talk about. So, the reality is that risk-less asset clause that people generally benchmark to, are the bonds issued by the government and in the US you will hear a lot about US treasury bond yields and India, how the RBI rates are how it affect so, those are typically what the ‘risk-less’ ones are.
Valuations really start coming in and those asset classes are really tied into macro-economics and how are country is doing and stuff like you know? Countries like Argentina are defaulted, Venezuela are defaulted, Zimbabwe has crazy inflation, I don’t even know if they have bonds trade but they maybe giving you 100% returns but the risk is so much higher. So, valuations really start coming in, in the context of the risk you are taking and irrespective of what people say, the reality is, there is one theory on which everything is valued, but nobody talks about it as much, it is called and this is where we have to get a little bit pedantic, it is called capital asset pricing model (CAPM), and underlying that, people can Google it, look up the formula and even post the formula, i.e.
Expected Return = Risk free rate + Beta * Risk premium
Where, Risk premium = Return expectation from the asset – Risk free rate and Beta = measure of risk in the asset.
But underlying that is to say there is a risk free return that you can get by investing in something very, very secure, like a government issuance of bonds and then as you start taking more risk, depending on the risk of the asset class, you want more premium to that return. So, if you invest in government of India bonds I don’t know the rate but let’s say six-six and a half percent. If you are investing in a bank FD your rate has gone up you want eight and a half percent, if you are investing in let’s say an early stage company you maybe want 50% percent, if you are investing in the stock market, you maybe want 15-20%. This is actually the theory and the underlying theory of that is, that essentially you are looking to generate, some kind of return, the return ultimately has to be cash flows and therefore, there is a model called discounted cash flow and that is really the present value of that is what the value of the asset is, the risk you are taking determines how much you discount it back the IRR -now all of this sounds very pedantic, it is, but that is the theory and that is the right theory. So, it is essentially capital asset pricing model which determines basis the riskiness of the asset what return you should be getting, and then depending on the asset, technically speaking even the companies we invest in, we should be projecting out their cash flows in the future and discounting them back at that return, that’s the price of that asset, that is the theory. It’s just not possible to use it because if I am investing in a very early~ and we’ll come to some of that. So, this is the first principles of how things are valued, and because this has a tendency to become GIGO ( garbage in, garbage out) because everything is an assumption, I don’t how to measure the risk of the asset, I don’t know how to measure the future cash flows, I don’t know so many things that it ends up becoming an excel spreadsheet which will potentially churn out garbage depending on the inputs. Now, in certain businesses it can be done. So, what happens is, people start moving- the best investors understand this theory and then they start moving away from this theory to second principles after understanding it and saying “what are the best proxies for it, right?”
Avnish: So, let’s go to public market stocks, what is the first proxy people pick price to earnings ratio, right? now interestingly a lot of people don’t thought about it this way but price to earnings if you reverse it, is earnings divided by price. That is like the interest rate you are getting. So, if a stock has a price to earnings of 20, the interest rate I am getting is 5% because it is 1 by 20, if I am getting 8% in the market, should I be investing in that? That is where things start becoming complicated because then the market says “what is the growth of this?” and stuff like that.
Another very simple thumb rule that people use, is in real estate which by the way no one clearly uses in Bombay because no real estate would sell here. The simple thing is, if I am buying a real estate asset and I rent it out, what is my yield on that? On the amount I have put it, how much is the rent divided by price I have paid? in Mumbai, in the city of Mumbai between residential real estate it is 2-3%, how much is the interest rate on the loan I am paying for it? 9-10%, should I be buying it? No, but yet people buy, it is an emotional decision, but it does not make sense on first principles. Commercial real estate in Mumbai six and a half to 7 % yield, makes a little bit more sense because you can actually get loans and there are deductions. So, these are how people come up with proxies to value assets because this first principle is not possible to always put down on a spreadsheet but that does not mean that the first principle itself is not correct, people also say “hey, this company~” so, in the stock market people or even in our business people say “This company sold at so and so multiple” So, they use that as a proxy so the next company will also sell on that multiple. Well, not really, you know whatever the fundamentals, people sometimes don’t think that through. So, like I said what is the takeaway? The takeaway is that there are risky assets, so before we talk about valuations, valuations are in the context of investing, investing is in the context of risk and return, and the risk and return is essentially ~ the theoretical model of that is a model that talks about how one should model risk and how one should model return which is around discounted cash flows but because those are very hard to model although in some businesses people do which are generally mature businesses, people tend to use proxies for it.
Salonie: Okay so that tells us what forms the basis of valuations in terms of first principles that people go by, but if we had to go slightly deeper this would vary as per the sector or the industry of the business that one is valuing. What are some of the thumb rules or principles to be conscious of as an investor and are there any that are specific to an industry?
Avnish: So, what is the challenge in the VC space? The challenge in the VC space is number one, companies often not making money, number two they are sometimes growing very very fast, I mean imagine if you had to price Facebook using this model or Google using this model when they were just starting out, today they are 700-800-900 billion dollar companies. Apple maybe, had some cash flows. Third, they are often illiquid, you can’t even you know trade them, so it becomes very very hard to figure out how does one value these businesses. So, we said we’ll deviate from first principle into thumb rules so people deviate from first principles even more. So, for example in e-commerce people use price to gross merchandise value, as a proxy. In enterprise businesses people us price to revenue as a proxy, in financial services people use price to book as a proxy so on and so forth. In capex heavy businesses people use price to EBITDA as a proxy. Now, here is what happens, and where the mistake happens, people think that these are new valuations techniques, they are actually not new valuations techniques, they are proxies for an underlying first principle’s valuation technique because you don’t have all the data to be able to apply that first principle’s valuation technique because the business is not profitable, I am not able to do a price to earning’s multiple, because the business’s cash flow is uncertain I’m not able to do DCF, but all of this we backed into from that same underlying first principle which was the capital asset pricing model. So, it is a little bit of that tail wagging the dog situation where the proxy becomes the rule, right? And the proxy becomes the original approach that is where all the mistakes happen, So, let’s take a couple of quick examples for people because these are proxies we moved already from present value of discounted cash flows to saying let’s use PE’s as a multiple, then the second thing I told you was, PE’s are in the context of growth, another thumb rule that people have is, peg multiple of one. If a stalk is growing at 30%, if a company is growing at 30%, the PE should be 30. Now, If I am valuing a company that doesn’t have E I can’t give a P so now 30 times 30 the market cap of the company ~ because PEG is one, price earnings to growth is one, 30 times 30 is 900 of a company that is making INR 100 of revenue, that is a 9 times sales multiple, so therefore, people say, okay if I believe this margin structure in the end state of this company, because I don’t know the current earnings, I will value it at 9 times sales
Avnish: The problem then becomes, often happens, people start using 9 times sales without thinking what the margin structure is, without thinking what the growth rate is, right? So, again the proxy starts becoming the rule, the tail starts wagging the dog, that becomes the actual way of valuing things.
Another example I’ll give you very quickly, I said high cap-ex businesses are valued on EBITDA because why? They have earnings but depending EBITDA as a concept takes out the~ how much debt and how much equity you have in the business and because it is a cap-ex heavy business, it likely has a lot of debt. So, people want to look at the operating earning of their business which is EBITDA. Now, let’s say people say, businesses growing 20% and 15 times multiple now what is that on price to sales? 15 times 20 is 300 that’s a 3 times price to sales multiple. So, like I said these are all proxies to the same underlying methodology one should look at a little bit of~ first of all, recognise that these are proxies and then look at a bunch of data to actually zero in on which proxy is right to use in a particular setting. Now, one caeat actually let’s come to price to GMV the most abused one in e-commerce, you know I have heard and I have seen companies valued at 6-8-10 times GMV. GMV is not revenue, revenue is, GMV times take rate, take-rate is typically the commission that the site is taking, it can be 10%, it can be 20%, now those are two very different numbers. So, if I believe, we discussed earlier that businesses can be valued at nine times sale, right? Or nine times revenue so, what does that mean in terms of GMV, if the revenue is 15-20% of the GMV that means that the business should be valued at 1.5 to maybe two 2 times approximate numbers of GMV, yet people talk about 8-10 times, again, why? The proxy becomes the end state. One is not thinking through, a business with take rate of 10% should be valued at half the business of a take rate of 20% , it gets more complicated than that but that is really the way to think about it.
Salonie: Right, so what you’re saying is that which proxy to apply in which setting is obviously key in helping you determine or at least brings you closer to determining the accurate value of the company. But coming back to my first question, how does one trule make sense of valuations?
Avnish: I thought that’s what I’ve been doing clearly I haven’t been answering the question well, so look I understand. I think you know the objective so far was to build some theory behind the practice. I think we have talked about some of that. I want to make one or two caveats before we try to actually hit the questions. So, we talked about various multiples, now the multiples are all in the context of a particular company and a particular quality of business. So, the margins of the business really matter and people should be thinking about these multiples in that context, there is this concept of pre-money multiples versus post-money multiples, often investors take whatever is convenient because if I’m doing a multiple on earnings and I show post money by definition that’s going to look bigger so people go to their investment committees with pre-money but the reality is that the next investor who’s coming in, that post money multiple has become a pre-money for them. So, lots of gamification happens unfortunately in our industry, the biggest one is in Fintech, where the only place where the price to multiple looks lower on post money basis is in Fintech. So, wherever it looks lower is where investors pick it, whereas what they should be doing is, wherever it looks higher is what they should be doing because that is what they are really paying. So, in Fintech people do this ponzi scheme of post money price to money book and stuff like that and that is a topic for another day but I think the context of how we use these multiples is very important. Now, I had mentioned earlier that valuations and coming back to finally trying to answer the question. Valuations are first of all, like I said in a context. Second of all, valuations are actually not GIGO.
The people who are some of the best investors in the world are very valuation sensitive, very price sensitive and price and value are as Warren Buffet says it very well, they are not the same thing. First, figure out what the value is, price is what is trading between people, and then try to get as close as possible to the value or below the value . Now, in our business sometimes that doesn’t always work. Multiples are driven by sentiment, value is the fundamentals. So, that is the core of how one should be doing this.
Now, I think the difference of how this works really in our business or in general is that the best investors spend time thinking about value versus price. You go to a meeting with an entrepreneur often people will say what you want or what is your price, what is your expectation? what is your valuation expectation?. Valuation expectation is not the value and what happens is people start confusing price and value like I said earlier so how should we be using it? We should be using it by recognizing that the value of a business is very different from the price of that business and ultimately that business is going converge, the price will converge to the value. Now, when you are at very early stages of a business it is for us to figure out how do we think future backwards? How do we think that at exit most likely the price and value will converge. So, for the sake of argument let’s take the e-commerce business as an example. If I have paid 6-8 times GMV believing that it’s a multiple of sales and that’s what the market is doing and that’s what everyone is doing. Five, seven, eight years down the road when that business is mature it is going to be valued let’s say one and a half times of GMV. That business has to grow many folds more because my multiple is about to contract, by call it 60%, 70%, 80%, depending on what I paid. I think that context keeping in mind is important in our business. So, there I would say VCs do think about that. I would also say that it varies by stage, if I’m at a very early stage investor or let’s step back - I would say think of it as three rules: one it varies by stage. If I am at seed or series A I think about ownership, how much capital I’m giving the business, how much runaway will it have. If I’m at series B I start thinking a little more about how this business will scale and what will happen. If I am at the later stages, I’m really thinking exit backwards. So, the second rule, first rule varies by stage, second rule think exit backwards when this business is ultimately sold what multiples happen in the market? Go through the history of the market, look at through cycle multiples and then go to what that value will be and then come backwards. Given where we play we generally can’t project so far in the future so, we typically say with our capital how far will this business go? What in that stage in a base case, not in a delusional case will be worth and then try to back into what we think the current value is.
Salonie: And what would your third rule be?
Avnish: Ha, the reason I didn’t say it, uhm so, the third rule is that real money is made when these rules are broken. So, the reality is, the problem is that the exception starts becoming the rule. So, ultimately one has to break some of the rules and somebody I really respect in this business said it really well which is, first, understand the rules and then break them. I think what I observed is that often ~ and by the way, typically there will be five, six, seven rules or fundamentals of first principles, you break one of them or two of them not all of them, right? So, I think the best investors in the world a) understand the rules b) understand which rule they are breaking and c) do it only a few times in their lifetime, not a few times in their year and Warren Buffet is a great example, I mean he is the ultimate, understands every rule of investing, when the financial crisis happened, he had been tracking Goldman sacks for I don’t know how many years and he put in 5 billion dollars over a 24 hour period, right? He broke a rule, who does without any diligence 5 billion dollars but you know he had been tracking the company for a while, but they don’t do it every day and then they don’t start believing that they are so good, they realize that they have broken a rule and they have done that as opposed to believing that breaking the rules is the rule. That’s the bottom of it.
Salonie: Got it. Thanks Avnish. Thank you for listening and you can find the transcribed version of this podcast on www.matrixpartners.in. You can also follow us on Twitter and LinkedIn for more updates.