Wednesday, January 30, 2013

Estimating Value: Part 3 Advantages and Disadvantages of Multiples in Valuation



Two previous posts have talked about the use of discounted cash flow in valuation and the use of multiples in valuation.  At the end of the multiples post I promised to return to the advantages and disadvantages of multiples.  That is the subject of today's post.


Advantages and Disadvantages of Multiples in Valuation
 The main advantages of multiples are that they are relatively easy to use, are based on actual market transactions and can provide a useful ballpark for estimating value. It takes a known quantity for a firm like earnings or book value and converts it into a proposed price for the firm.  The problems associated with multiples are many, starting with the difficulty in finding comparable and timely comparisons.  The multiple approach also presumes that the prices paid for competing firms or their shares of stock were realistic in the first place!  Moreover, the average multiple for a set of transactions disguises the wide range of differences that can exist within a given industry. It is not unusual, for example, to observe an average multiple of 15 times earnings for an industry with a range from 10-20 times earnings representing actual transactions.  Selling your firm for the average of 15 times earnings when you could have received 20 times earning is obviously not maximizing shareholder value. 

To illustrate the wide range of multiples within various industries, consider the Table below taken from some work I developed using Value Line Investment Survey.  The table shows the mean, median and range of price earnings ratios for ten industries during the summer of 2001.  Of particular interest is the extreme range of the multiples within a given industry.  Note the ratio of the range shown in the far right column.  There are only two industries where the range in values from low to high is less than 50% of the median value; both of these industries involve just a handful of firms.  


Table 1: Range of price earnings ratios across industries
Name of Industry# of Cos.MeanLowMedianHigh  Ratio of range     (high – low) to median value
Aerospace/Defense1615.26.314.827.5143%
Auto Parts Industry17178.414.730.2149%
Bank Industry3017.913.216.732.2114%
Beverage (soft drinks)824.321.823.827.424%
Hotel/Gaming Industry1617.311.417.131.1116%
Medical Services2818.86.31929.6123%
Newspaper Industry1331.119.130.348.698%
Petroleum (Producing)9127.710.624.2156%
Retail Store Industry2020.97.521.143.3170%
Tobacco Industry611.38.811.313.441%
Source: Value Line Investment Survey – Issues from 7/6/01 through 9/28/01.


Consider the Auto Parts Industry.  The range is from a low of 8.4 to a high of 30.2.  Can you confidently price your company at any of these valuations (or the mean or the median) without more fine tuning?  I think not.  

Moreover, multiples are based on rules of thumb, often learned through the experience of practicioners.  But consider just a few of the factors affecting valuations: economic conditions, consumer tastes, the existence of war, technology, alternate products, complementary products, inflation, etc.  etc. The list could continue for quite some time.  As any of these factors change, so could the multiples.  Even the relative ordering of multiples within industries will change over time (inflation or the price of oil affects industries differently).  This set of multiples measured today would vary considerably.

Also remember the types of multiples shown in the use of multiples  [from Snowden (Table 1)].  In many cases, a range of multiples is provided and often, the range of multiples seems so large as to provide little help in estimating a firm’s value.  

Indeed, analysis of the multiples existing within any particular industry is likely to produce a very large range.  The analyst needs to make subjective adjustments recognizing differences between the firm of interest and the comparison set.  This involves recognizing the current and future earnings power of the companies, reflecting current market conditions, and identifying other important differences between the firm and the comparison set.  At the extreme, recognition of these differences leads to an analysis reminiscent of the detail inherent in a discounted cash flow approach.  

This is not to say that multiples are without merit.  On the contrary.  Multiples are just one other means of valuation with their own advantages and disadvantages.  They also serve as boundary checks on methods like discounted cash flow valuation.  Finally, many analysts that use discounted cash flow to value the next several years of earnings will still use multiples to estimate terminal values.

In the next valuation post, we'll conclude with a look at a related method of valuation: comparables.

Monday, January 28, 2013

Dell LBO: The Beat Goes On


Part 2 of our discussion of using multiples in valuation will appear Wednesday.  We wanted to get this timely post on Dell out today.

R

-----

The potential Dell LBO continues to develop. Microsoft is considering a $3B convertible preferred stock investment. This would greatly strengthen the outlook for the transaction. Besides improving the capital structure, it provides a deep-pocket strategic partner. This is crucial to provide comfort to debt providers as a secondary repayment source should Dell experience financial problems.

Microsoft has a strong interest in seeing Dell, a major customer, survive. An out-right Dell purchase might pose concerns with some of Microsoft's existing suppliers and Window's allies. The joint venture with Silver Lake, the lead PE firm, provides a toehold investment with the implicit option for an additional follow-on acquisition. Should Dell falter, then Microsoft could elect to support Dell by increasing its ownership under the cover of protecting its investment. If Dell prospers, then Microsoft could serve as a strategic exit partner. Bottom line, they obtain an inexpensive option for a subsequent acquisition, reduce initial supplier and allies concerns and avoid having to consolidate Dell on its balance sheet.

Next to consider is the value transfer from Dell's bondholders to the transaction.  Like most investment grade bondholders, Dell is currently rated A2/A- by Moody’s and S&P respectively.  Dell's bondholders have minimal covenant protection and cannot force Dell to repay them once Dell goes private. Thus, Dell avoids having to refund the low rate bonds with more expensive buyout debt.  The existing bondholders continue to receive an investment grade coupon on their investment, which is now non-investment grade-most likely BB post close. Worse yet, the bonds will become effectively subordinate to the buyout debt. Consequently it comes as no surprise that Dell's bonds have plunged in price since the LBO discussions began.

I still question the wisdom of a highly leveraged Dell trying to adjust to its strategic challenges in a maturing and highly competitive industry facing substantial competitors who are unburdened by high leverage. Nonetheless, the potential for financial engineering related magic may just make the transaction possible even if it remains unwise. The Dell Silver Lake guys are good.

More to follow I am sure.

J


Friday, January 25, 2013

Estimating Value: Part 2 Using Multiples

Following up on two previous posts, ( Value is Estimated, Price is Paid and Estimating Value: Part 1 DCF ) we begin a multiple post discussion of the use of multiples to estimate value.  We use multiples all the time in our daily lives.  As just a few examples consider driving:  miles per gallon, miles per hour, price per hour of parking.  Consider college: price per class,  price per credit hour, average price per book.  Consider housing: price per square foot (or meter), price per kilowatt hour (for heating).  Consider advertising: price per (internet) hit, price per name (for mailing lists), price per column inch (for print advertising).  Consumer products are standardized as price per unit (often an ounce, etc.)  And certainly our wages are expressed in multiples: salary per year, dollars per hour.

Multiples are pervasive.  So it should not be surprising that multiples are used in business valuation.  Note that in each example listed in the first paragraph, the multiple was expressed as a factor of a base (the item after per).  The base should be something meaningful to what we are estimating.  In business valuations, some common bases include: earnings, book value, EBITDA, sales, cash flow, growth, customers and other items.

In business valuations, we choose an appropriate multiple for our industry, find the relevant value for our company (by looking at the average or weighted average multiple of comparable companies), and apply the result to the appropriate base value for our company.

As an example, consider the price earnings ratio.  When we are buying a company (or a fraction of a company, i.e., a share of stock), we are essentially buying the future earnings stream of a company.  It makes sense to use earnings as a meaningful base from which to project price.  So let's say we estimate next years earnings for our firm to be $3. per share and we find a set of companies similar to ours with an average price earnings ratio of 10.  The implication is that our company is or will be worth $3 x 10 = $30. per share.

Similar calculations would be used for other bases (book value, sales, etc.)  In some cases one type of base is more meaningful for a particular industry.  In other cases, analysts might average several valuations resulting from various multiples or weight the valuations according to factors relevant in the particular situation.

A very partial list of some multiples used in practice include:

  • Price to earnings
  • Price to book value
  • Price to sales
  • Enterprise value (i.e., equity plus debt) to sales
  • Price to EBIDTA

etc.

Practitioners have also evolved elaborate variations on strict multiples. (Note: I'm not commenting on the merits of these multiples here or the superiority or lack of it in the multiples listed below.  I will, however, discuss advantages and disadvantages of multiples in general in my next post.)   Consider a list of industry multiples provided by Snowden (1994) and shown in Table 1 below.  The items being multiplied vary by industry.  Many also include specific adjustments for inventory or equipment.

Table 1:               Industry Multipliers
Industry
Multiple

Travel agencies:
.05 to .1 X Annual Gross Sales
Advertising agencies:
.75 X Annual Gross Sales
Collection agencies:
.15 to .2 X Annual Collections + Equipment
Employment agencies:
.75 X Annual Gross Sales
Insurance agencies:
1 to 2 X Annual Renewal Commissions
Real estate agencies:
.2 to .3 X Annual Gross Commissions
Rental agencies:
.2 X Annual Net Profit + Inventory
Retail businesses:
.75 to 1.5 X Annual Net Profit + Inventory + Equipment
Sales businesses:
1 X Annual Net Profit
Fast food (nonfranchise):
.5 to .7 X Monthly Gross Sales + Inventory
Restaurants:
.3 to .5 X Annual Gross Sales, or .4 X Monthly Gross Sales + Inventory
Office supply distributors:
.5 X Monthly Gross Sales + Inventory
Newspapers:
.75 to 1.5 X Annual Gross Sales
Printers:
.4 to .5 X Annual Net Profit + Inventory + Equipment
Food distributors:
1 to 1.5 X Annual Net Profit + Inventory + Equipment
Building supply retailers:
.25 to .75 Annual Net Profit + Inventory + Equipment
Job shops:
.5 X Annual Gross Sales + Inventory
Manufacturing:
1.5 to 2.5 X Annual Net Profit + Inventory or  .75 X Annual Net Profit + Equipment + Inventory (including work in progress)
Farm/heavy equipment dealers
.5 X Annual Net Profit + Inventory+ Equipment
Boat/camper dealers:
1 X Annual Net Profit + Inventory + Equipment
Professional practices:
1 to 5 X Annual Net Profit


Source: The Complete Guide to Buying a Business, Richard W. Snowden AMACON, New York, 1994, pp. 150-151.


These were given at a particular point in time and learned from actual experience.  As you will learn in our next post, they are unlikely to be relevant today, but you can see variation in multiples suggested with various adjustments for different industries.  

From all of this detail you might assume I am a huge fan of using multiples in valuation.  I am not.  Don't get me wrong, multiples are an important part of our valuation toolkit, but they have distinct advantages and disadvantages.  It is important to understand these attributes before using multiples.  We'll cover that topic in Monday's blog.

All the best,

Ralph



Wednesday, January 23, 2013

Estimating value: Part 1 An Overview of Discounted Cash Flow

Previously, we've written that Value is estimated, Price is paid.  Today we  start to discuss some of the many ways to estimate value.  Each method has advantages and disadvantages.  None is perfect but taken together they provide a meaningful framework for estimating value.

Some of the major techniques used to estimate value include:



  • Discounted cash flow
  • Multiples
  • Comparable Transactions


In today's post, I'll provide an overview of discounted cash flow, returning to multiples and comparable transactions at a later time.  We'll also ignore liquidation value and book value for this post.  Liquidation value is useful when you are anticipating dismantling a company and book value is, well, historic.  The book value of any asset doesn't necessarily reflect true value and is often dramatically different.  That said, there seems to be some  information in book value.  Multiples of book value, for example, are often used in valuation.


True or intrinsic value

So we mentioned 'true value'.  It is often called 'intrinsic' value, and yes, it exists in the eye of the estimator.  The true worth of an asset, of course, is what someone else will pay for it.  But in valuing our company this is unknown.  How can we get estimates of intrinsic value, estimates that inform a meaningful selling price?

DCF in practice

Let's start with discounted cash flow.  Simply put, the value of any business asset is the stream of cash flows it will generate throughout its life, expressed in today's dollars.  We generally assume that businesses, and the equity that represents ownership in these businesses have an infinite life.  Products follow a life cycle.  Companies that produce them can last indefinitely by continuing to adapt, replacing old products with new ones.  Now, we can't meaningfully think about cash flows at year seventeen, let alone at infinity so we often use a shorter term investment horizon.  Let's take five years and assume (probably artificially) that we sell the company or stock at the end of this period.  Whether we do is immaterial, it just helps us conceptualize the issue.

So in the diagram below we have two cash flow streams, one stretching to infinity and one truncated at five years.   The present value of each of these streams is the current worth of the company.



The equation describing the discounting process (converting the cash flows into today's values) is shown below.

Terminal value
Note that in the arrowed diagram and in the equation the cash flows stretching from year six to infinity are replaced by SP5, the terminal value.  This is the assumed selling price in year 5.  How do we estimate this selling price?   Two standard techniques are: a) a constant growth model and b) price multiples.  The first technique is a mathematical reduction of assuming cash flows will grow at a constant rate forever.  The second is based on the valuations of related companies.  Both methods have some merit.  I'll cover multiples in a separate post, but let me address an often voiced concern with assuming 'constant growth'.  Sure, no company is likely to actually have constant growth, but as we noted above, it becomes impossible to estimate growth precisely at some date in the future.  What is your estimate of growth in year 17 ??  Our choice is to either ignore the future or incorporate our best guess.  Constant growth does the latter.  Also, commonly used multiples (of earnings, book value, etc) have built-in but unstated assumptions just as heroic.  A word of caution: probably 70-90 percent of a firms value will occur after year five.  Be very  careful with that estimate.  


Advantages and disadvantages of DCF
An advantage of the discounted cash flow technique is that it can be applied to any business from purchasing a motel in Orlando to a photography store in Crested Butte to purchasing General Motors. In each of these examples, the value of the business to the owners is equal to the present value of all the distributions the business will generate. 
It is sometimes argued that discounted cash flow is difficult to understand.  To the contrary, the mechanics are easily applied using spreadsheets or financial calculators.  The intuition behind discounting is also easy to understand: money received in the future is worth less than money received today. The intuition behind the rate we use is also straightforward: We should discount cash flows at the appropriate opportunity cost. In other words, evaluate this project at the rate we could earn on projects of similar risk.  Our answer to the question: “What rate could we earn elsewhere with the money at this risk level?” answers the rate question.  And since higher discount rates (associated with riskier ventures) lower the present value this method adjusts for risks.

It is also asserted that the DCF method requires too many assumptions.  It is true that forecasting the amounts received in the future and estimating the opportunity cost of funds require challenging assumptions.  Nevertheless, all valuation methods utilize assumptions.  With some techniques, however, the assumptions are hidden in the apparent simplicity of the process.  As a consequence, users are often making implicit assumptions without even being aware of it.  A major advantage of discounted cash flow is that the assumptions can be quite explicit.  The ability to fine tune projections of cash flows to recognize individual firm characteristics is a tremendous advantage. Moreover, most alternatives to discounted cash flow, such as using ad hoc rules of thumb to value an asset, do not allow for easy updating. Rules of thumb that held in the past may quickly become obsolete if some underlying factor of the economy (e.g. inflation) changes. Discounted cash flow with its component parts expressed in a spreadsheet allows for immediate revisions.

In a subsequent post, we'll take a brief look at multiples and comparables.

Monday, January 21, 2013

The Dell Stock Repurchases Program - Hmmmmmmmmmmm


Ralph and I have an on-going debate on the merits (Ralph) and demerits (me) of share repurchases. My point is not all repurchases are the same. Some may be value enhancing. Others, however, are not. They can be used by executives to manipulate earnings per share (EPS) to improve their option values.

Floyd Norris' January 18,2013 New York Times article on the misuse of repurchases at Dell provides a cautionary tale for shareholders. He analyzed Dell's long running approximately $ 40 B program. He concludes over priced repurchases benefited corporate executives at the expense of long term shareholders. Michael Dell, Dell’s founder and largest shareholder, was a substantial option recipient. His gain is estimated at over $650 MM. Executives would profit if Dell stock price increased following a decline in the number of outstanding shares post repurchase compared to a dividend. This encouraged repurchases to offset potential stock price weakening by increasing EPS as operations began to mature. This occurred despite the negative impact on long-term non-tendering shareholders. This occurred because the repurchases, being over priced, transferred value from remaining shareholders to those tendering. Michael Dell's losses were offset by his option gains, which were unavailable to non-executive shareholders, plus the gain on any shares he tendered. Dell paid an average repurchase price over the years of $ 19 per share compared to the current price of $ 12. A general observation is firms facing a challenging operating environment who have stock option programs may be prone to questionable repurchases.

The current low share price due to concerns over the firm's business model has increased the interest in taking Dell private. Michael Dell is rumored to contribute his shares into the potential LBO. This raises conflict issues as to who he represents-the firm's shareholders or himself in the buyout negotiations. Given his record on the repurchase program we can only guess hmmmmmmmm.

I promise this will be the last post on repurchases for a while. There will be more posts on Dell as the story develops.

j