Reaching a Conclusion of Value: Reflections on the Use of Market Data
Posted on 09.12.14
Most of us valuation analysts love dealing with the details of financial analysis. We immerse ourselves in the details of items like determining the precise rate of return, an exact future growth rate, or the precise method of considering the timing of an expected future cash flow. What sometimes gets lost in the details of this analysis, though, is that our job is to determine a value that is representative of the price that would occur in the relevant market. In the end, we are not supposed to come up with the best theoretical financial analysis but, rather, the price that is likely to be realized in the market. Heated discussions about things like the exactly correct equity risk premium or the appropriateness of the size premium are important and intellectually stimulating but they should not deflect attention from the proper emphasis on answering the question of essence: Is the conclusion of value the likely price that would be agreed to in the market? While I do not discount the importance of financial analysis considerations, it is important that the objective of the financial analysis is to determine the price likely to prevail in the market.
Not only is it common sense that a fair market value should reflect the market but it is also a requirement for federal tax fair market value determinations. Federal Estate Tax Regulations Section 20.2031-1(b) provides (among much other guidance), “Nor is the fair market value of an item of property to be determined by the sale price of the item in a market other than that in which such item is most commonly sold to the public, taking into account the location of the item wherever appropriate.” I have always taken this as a directive that I had to consider the effect of the specific market and not some general notion of a market. This means that for a private company, we have to specifically determine the price in the market for a private company or an interest in the private company.
This is all a laudable enterprise, but showing that the concluded value is the likely market price for a private company is a uniquely challenging task. The best theoretical way to do this would be through the use of exactly comparable market data – a sale of this exact interest close in time. Of course, this is almost never available for the private companies most of us value. So we fall back to the next best way, which is market data that has a “reasonable basis for comparison.” (These words are directly from ASA Business Valuation Standard V.) Unfortunately, this kind of data is also difficult to come by. This lack of data doesn’t lessen the obligation to determine a price grounded in the market, but it does mean that a methodology short of direct comparison needs to be used.
In the absence of specific market data, one way to incorporate the effect of the market is to duplicate the process used by investors in private companies. I take comfort in the fact that transactions of private companies frequently take place and the buyers and sellers of these companies face all of the same information difficulties that we do.
The way that I think the market works for most private operating companies is through a process which involves a combination of the income approach and the market approach. I am influenced in my thinking by a survey effort I undertook with a number of my associates many years ago when I was a partner at a big accounting firm. We called a number of clients and others who were active acquirers of businesses. This wasn’t a large survey but it did involve in-depth discussions with private equity investors. What I found was that the best way to characterize the market price was through a process. The price most of the buyers were willing to pay was generally based on consideration of multiples prevailing in the market plus consideration of an adequate internal rate of return or present value.
The multiples mentioned were MVIC/EBIT and EBITDA in an approximate range of 3-7 and most of the investors did consider that the range changed over time based on the state of the market. The internal rates of return for private equity investors were in a range of of 25 percent to 40 percent and the internal rates of return for strategic investors were in a range of 10 percent to 20 percent. The return required in a specific deal also considered the general returns believed to be demanded in the market at the time of the deal. The multiple was applied to a representative income number that was based on the buyer's analysis of where the company was today, where it was headed, and what that buyer could do with it. This was all an iterative process in which the investor was assessing whether the multiple and the rate of return were too high or low given the expected future opportunity and risk.
This survey effort is now about 25 years old, but I don’t think that the general conclusion has changed much in the way the process is carried out or the range of multiples or rates of return. What leads me to say things haven’t changed much are the survey results from the ongoing Pepperdine Private Capital Markets Project. This survey shows multiples and rates of return similar to the range discussed above. Also, the Pratt’s Stats Private Deal Update shows EBITDA multiples for the smaller companies it covers in the 2-4 range.
So my overall approach in valuing an operating company has been to try to replicate the process that the private equity investors told me about. I had to turn it around a little bit, but the results are the same. I carry out an income approach that takes in everything that I think an investor would consider and then use market data to help me determine whether this makes sense.
My idea of how the income approach should look is that it should represent what I believe an investor in the market would do. The projection of income, the projection period (if it’s a dcf), the rate of return, and the assumed amount of debt are all based on my perception of what would prevail in the market. The emphasis in the income approach is to make it reflect what market participants would do. Theoretical considerations are good but only if I believe that market participants would use them. The Pepperdine survey data is of some help in getting an idea of what market participants are thinking.
The market data I use is based on the best data available for the particular case. I always consider as best I can the current state of the market. Some sense of this can be obtained from Pratt’s Stats Private Deal Update (published periodically by Business Valuation Resources) and the Pepperdine survey data. Also, I always look at Pratt’s Stats and Bizcomps for specific guideline companies. A time or two, I’ve found a company of the right size, line of business and close in time. I wish it happened more often, but this is a rare occurrence. I seldom find specific transactions where I can convince myself that there’s a reasonable direct comparison without a stretch of the logic. The quantity and quality of information about the companies included in the databases makes direct reliance on specific transactions problematical. A very significant issue is that the investor’s assessment of important qualitative issues such as the risk and opportunity can’t be gleaned from the data.
Also, I look for reasonable public guideline companies to include in my market data if the subject company is large enough or has appropriate growth characteristics for smaller companies. It is an even rarer occurrence that I can find such a company with a convincing reasonable basis for comparison.
The way I use the market data to determine that the income approach appears reasonable is to consider the factors that affect the price. Some of these are the size of the business, the debt-bearing capacity of the company, stable history of cash flow and likelihood of continuation, risk of the company, expected future growth, competition and barriers to competition. Both the Private Deal Update and the Pepperdine survey show multiples by size but beyond that I don’t have a way to quantify the effect of these factors. It is a subjective adjustment. Although this is a judgment call, it seems obvious to me that a company with good marks in all of these factors would command a higher multiple than one not so blessed.
Even though I seldom find private transactions or public companies that are good direct comparisons, if there are a number of observations available, the range of information from these sources can sometimes be extremely useful in showing a pattern and range of pricing. In some industries, I may find that multiples of book value, sales or EBITDA provide a range of pricing that should be considered. Also, this market data can be useful in highlighting a case that falls outside of the general methodology I have laid out above.
I've also got to point out that there are a number of situations that fall outside of the general scheme I've discussed. Examples include start-up companies, troubled companies, and industries with unique factors that lead to a different pricing methodology. These are usually cases where the classic income approach may not be the primary approach.
In the ASA Business Valuation Standard on Reaching a Conclusion of Value, the heading for Paragraph III now reads “Selection and Weighting of Methods.” I was the chairman of the committee that originally wrote this standard and we did not use the term “weight.” Instead, it read “weigh.” This was not a typo but was intentional. It is too bad that it was changed in a subsequent revision because it changed the meaning. The original idea for the term “weigh” was that when you reached a conclusion you would not always use a mathematical weighting, but rather you would consider all the approaches together and weigh the results of different approaches and methods to reach a judgment on the best conclusion. The general scheme I have discussed above is just such a weighing process where an income approach and market data are considered together to formulate the final conclusion.
It is our task as business valuators to determine the likely market price that we believe would prevail in the market. It would be pleasant if there was hard and fast market data that could be used to “prove” such a value, but this can seldom be done in a precise way. Unfortunately, the market itself isn’t based on such data and participants use the kind of information and the process I have discussed above to determine a price that is in line with their unique goals and perceptions. The best approach to determine the market price is to duplicate the process used by likely market participants and with assumptions that are believed likely to be used by them.