What risk-free rate to use?

I recently wrote in my summary of Grabowski article on the cost of capital that he now proposes to use a longer term average of risk free rates instead of the current risk free rates, at least until these rates return to their normal levels. So, here are some thoughts on this:

- True, risk free rates are extremely low right now. But does this mean we should just ignore market inputs into the risk free rates because we don't agree with them? Who are we to say that we know better than the market (which is composed of thousands/millions of participants) what the appropriate risk free rate is today?

- If Rf today is 3,4% and you use 4%, you automatically assume that the market itself is overvalued. But is this the assumptions that is required of you when estimating the market value? Furthermore, today you can invest in a risk free asset with 3,4% yield. Investing in equity should yield a higher return and this is reflected in risk premium add-ons. But by using Rf of 4% you basically assume that this same investor could today invest in a risk free asset and earn 4%. Which is simply not true.

- I think it all depends on the premise of value and subject of the valuation. When valuing a market value of quoted stocks you should use the input from the market. When valuing market value of significant minority or majority stakes you should use the discount rate used by potential investors. If you believe they base their discount rate calculations on long-term averages of risk free rates, then so be it. When valuing the intrinsic (fundamental) value of a company, then you should definitely use the risk free rate you feel most comfortable with, be it the current rate or the long-term average.

I'm not against making these adjustments to discount rate calculations. I'm just saying they should make sense and should not be done automatically, just because we don't agree with the market.

As argument goes, due to lower risk free rates we today end up with lower discount rates, which is counterintuitive considering the increase in the perception of risk. My answer is:

- the greater risk and the crisis itself should already be reflected in the cash flows;

- we should compare the cost of equity with the risk free rate (using today's inputs) and not with cost of equity as of a year ago. Back then you couldn't invest in a risk free asset at 3,4% as you are able to today.

Financial projections - and the effect of the crisis

I have already stated my belief that now more than ever (ok, at least more than a few years ago) DCF method is the way to go. However, current events have increased the uncertainty of cash flow projections, so here are a few pointers I think we should all have in mind when using the DCF to value a company:

1. Macro analysis should not be just a piece of sheet in the report being there for the report's sake!

2. Sector analysis should be done as thoroughly as possible. Check different data sources, make your own conclusions. Don't rely simply on trends and averages in the industry, but rather ascertain the position of the company within the sector. Is it well positioned to take advantage of the current situation, or will it come out on the losing end? 

3. It will take longer to get to the steady state, so the length of the forecast period should be extended; do not automatically settle for 3-5 year forecasts (as you shouldn't a few years ago either, but hey...)!!!

4. Since forecasts should be extended and since there's an increased uncertainty, we should think about increasing the number of possible scenarios. Instead of just optimistic/pessimistic scenario try several different options. The result might be a greater range of possible values, but that's perfectly fine. It's hard to pinpoint an "exact estimate" (oxymoron, I know) of value anyway, especially now.

What's the point of a valuation date?

Recently I've been in a discussion regarding a valuation of a company as of some date a few years back (it was for tax purposes). And someone (a BV professional) said "but we can't just pretend we don't know what happened after that date!".  That's exactly what we MUST do! The point of the valuation date is to have a sort of a cut-off date, a point in time at which we would like to know what the estimated market value of the company would be considering information we had at the time. For example, let's assume we're valuing a company as of 31st March 2009. Generally, today (it's May 2009) one wouldn't pay much detail to what's happened next as one shouldn't - and here there's usually no dilemma. Now, consider we were engaged to do this same valuation in 2012. A lot has happened since. We know how the company performed. Should this influence our estimate of the market value as of 31st March 2009? NO! At that time we didn't have the information we have now. Of course, market value estimate as of TODAY is (or - may be) different. But when valuing a company on a specific date you have to go back to that point in time and ask yourself "what information do I have now, how much do I estimate the market value of the company to be". Should this be any different whether you do it today or 10 years from now? No. Will it be more difficult to be objective and ignore the information you have 10 years from now? Definitely. But that's just something you have to deal with.

It's sort of like saying: if I had the information I have now at the time of the valuation date, would I have made a different decision? Perhaps. But I didn't and I couldn't! For example, when making forecasts - could anyone as of the valuation date (which let's say was 31st December 2006) have predicted the occurence and extent of the financial crisis and recession and their influence on cash flow forecasts? Since we know about this crisis now, would we have made different cash flow forecasts back then? NO, because back then we didn't know about this crisis.

I know for some people it's difficult to come to terms with it, but the valuation date has its purpose and an appraiser should only take into account the information he/she had on that date!

Recommended reading: Grabowski - Problems with Cost of Capital Estimation in the Current Environment

I just finished re-reading the article by Rober Grabowski titled "Problems with Cost of Capital Estimation in the Current Environment" (you can find it at: http://www.duffandphelps.com/sitecollectiondocuments/cost_capital_update_012909.pdf#page=1). I strongly recommend this article; one may not agree with everything written there, but it's a great food for thought and coming from such an eminent author - it's probably a "must read" for any BV professional. As for the contents - the title says it all. Just to sum it up quickly, here are some of the more interesting insights:

1. Stock market correction has been concentrated in stocks of financial and highly leveraged companies. So, traditional beta estimates provide us with wrong estimates of systematic risk going forward. Author suggests looking at a longer period and see how company's returns moved as compared to the overall market and see which period is the one from which to choose beta to use in our estimates of the riskiness of future cash flows. Author also suggests some alternative methods for estimating betas.

2. Currently risk-free rates are extremely low, leading to lower estimates of the cost of capital, which - in this environment - is counterintuitive. Author suggests using a longer term average T-bond yield (e.g. 4.5%) or a forward rate on T-bonds, at least until yields on T-bonds return to normal.

3. Equity risk premium (ERP) is cyclical. Author claimes long-term ERP is between 3.5 and 6.0 %. And during recessions it's more towards the upper end of the range. He suggests using 6.0% ERP, which can be expected to decrease in the future once economy rebounces.

4. Leverage: current D/E ratios are extremely high due to low market capitalizations. It's questionable also whether they are sustainable (WACC can be calculated using a changing capital structure). Furthermore, typical models of un- and re-levering betas do not work well in this economy, since they imply several assumptions (constant D/E, tax deductions on interest realized when interest is paid etc.) that are not realistic these days. So, alternative models should be used (e.g. Miles-Ezzell formula which also uses debt beta).

5. Financial distress is not incorporated into COEC calculation, so in case of companies in financial distress a premium should be added (e.g. one calculated by Duff&Phelps).

6. Value of tax shield: WACC should be calculated using a formula that takes into account the probability that income tax savings will not be realized in the period in which the interest is paid.

In the end it comes down to common sense - make sure you check the reasonableness of your cost of capital estimate. Duff&Phelps again provide data on historic equity returns based on companies' average operating margins and their variability.

As I said at the beginning, the article is great as it makes you think about the inputs you use in calculating the cost of capital. I agree with some of the propositions in this article, but might take issue with others in my further posts.

Is "Investment value" always higher than "Market value"?

No. It seems counterintuitive and most people automatically assume that investment value surely is  higher than the market value, since - in the end - it is investors that determine the market value, or should I better say the price.

Market value is - to put it in simple terms - the value that could be achieved, if the company was sold on the market to any (and not some specific) market participant. It of course depends on a number of factors, among which are also the number of interested investors etc. Now, assume you have an investor who already owns 100% of a certain company. They engage you to do a valuation and determine the market value of their ownership interest in the company. Then they also want to know what the investment value is, because they've heard that investment value is even greater than market value. However, investment value depends on assumptions put forth by a specific investor. Let's assume this specific investor has a plan for this company, on the basis of which they will earn very low returns, while investing a lot. Utilization of the assets of the company would be far from optimal. So, the value you arrive at is in fact lower than the market value, because market value is determined by what market participants observe could be achieved with the assets of this company, i.e. they could earn much higher returns. 

So, you go back to the client and explain that, ehm, investment value is lower than the market value because of the way you manage the company...oh joy, don't you just love these conversations?

Another example is that an investor is a company that puts great emphasis on the local environment and people and one of the values they hold dear is to provide jobs to the local population. Even though any rational investor would rationalize their business operations. And this is something that you can hardly put into figures. So again, investment value (for such an investor) <>

So, could we find a way to put this soft stuff, that some investors are so fond of and that actually are one of the driving forces behind their motivation to do certain transactions, into specific numbers and attach monetary value to it? Then perhaps we would see that investment value indeed is greater than market value, even for such an investor. Well, I can just say good luck to those who try and congrats to those who succeed (though I don't expect to see many of those).

Using the market approach to value a business in the time of crisis

Lately, when valuing private businesses in these times of crisis, I'm often faced with an interesting dilemma. Is the use of the market approach in these times reasonable at all?

1. Finding truly comparable publicly quoted companies is difficult these days. Many companies have internal issues that can seriously affect their value, such as financing issues, liquidity issues, focus on the market segment which is in steep decline, "wrong" customers (e.g. in automotive parts industry, where most companies are doing badly and some prosper),... Additionally, the company we value may have such problems itself. Perhaps the industry sales have on average dropped by 20%, while the company we are valuing experienced a 40% drop in sales. Market approach may work just fine (on average) in a stable or growing market, but in such an unstable environment it does not allow you to take into account some very specific factors that are very much present at this point in time. On the other hand, DCF method allows you to do just that.

2. Using comparable transactions (for valuing majority positions) from the past, when there was a different climate in the market, is also debatable. In the past prices have been driven up by the presence of many investors on the market. We had strategic investors in a buying mode and financial investors with heavy wallets, ready to compete with strategic investors on almost any transaction. Financing was also (more or less freely) available. Today, the situation has changed. Potential investors with financing in place are a rare bread nowadays, so it's a buyer's market. Using high multiples from a recent past does not make much sense now. 

So, where is the problem? Owners of companies are still under the impression of days gone by. Up till about a year ago they had offers on the table for certain amounts that may not be achieved today. So, when they see your valuation, they start waving those offers and asking you whether you've lost your mind. In most cases it's possible to explain that we're in a different environment now - they don't like it, may even disagree ("for us it's different"), but accept your arguments. But in some cases they disagree up to the point you start doubting your decisions. So you recheck your assumptions. And I have to admit that in certain industries transaction multiples just might remain high (e.g. media, energy), since transactions are driven not only by pure economic motives, but also by some "higher", "strategic" motives. But such industries are rare and investors on average rational (even more so in the light of this crisis).

So, as a matter of principle I am sceptical of using past transactions' data in this time of crisis, but I do a thorough analysis in each case to see whether it might make sense to rethink this position.