Fructivore Investments

As I stated last week, I have taken the view of James Montier that a low price/sales ratio is an unreliable indicator of cheap stocks, since it ignores profit margins and capital structure. I have similarly adopted his view that an astronomically high price/sales ratio is a reliable indicator of overpriced stocks, because at a certain point there is no capital structure or profit margin that will justify the current price. Furthermore, the degree of growth required to justify a company at that price would be almost impossible to realize. And it seems to me that with Red Hat Inc. (RHT), we have a company in such a situation.

Red Hat Inc. provides software support services for open-source Linux programs in the enterprise space, including product delivery, problem resolution, ongoing corrections, enhancement and corrections, new versions, and compatibility issues, as well as providing training. Their best-known product is the Red Hat Enterprise Linux operating system, and they also include a virtualization platform (which may explain some of the optimism surrounding the company; VMware trades at a similarly high valuation), as well as middleware tools. All their products are open-source rather than proprietary; their revenue, apart from training, comes from the support they provide, which is sold on a subscription basis.

The business model is an attractive one and seems to be effective, no doubt, but Red Hat’s valuation seems to come straight from 1999; they have a price/sales ratio of 9.96 and a P/E ratio of 85. Those figures have recently increased by a 10% reaction to the firm’s 2nd quarter earnings announcement that sales increased 20% year over year and operating earnings increased 23% over the same period. In the current economic environment, such a level of growth is indeed impressive, and the open source model is enhanced by an army of tinkerers inside and outside the company. All of that makes for a good story, true, but as Damodaran reminds us in The Dark Side of Valuation, prudent investing calls for all stories to be translated into numbers.

Ultimately, reported earnings were down from a year ago, which the company blames on amortization and stock-based compensation. Now, as it appears that Red Hat’s stock is very expensive, the company is probably better off paying its staff in stock rather than cash. However, from the perspective of a shareholder, dilution is still dilution, and so their removal of stock-based compensation to arrive at their pro-forma earnings seems a little like cheating to me.

Turning to the amortization, however, I prefer to eliminate depreciation and amortization and replace it with capital expenditures in order to get closer to determining free cash flow for the period. From that perspective, Red Hat claimed total depreciation and amortization of $11.7 million, and total capital expenditures of $9.7 million last quarter, which would increase quarterly earnings to $25.7 million. The comparable figure for last quarter would be $32.7 million, after making an adjustment of $3.7 million for the same reason.

However, considerable further adjustments have to be made to strip out the Red Hat’s non-operating income and expense and to address the issue of deferred revenues in order to enable a real history of earnings growth. It is a somewhat labor-intensive process, but necessary in my opinion, to arrive at figures that most readily permit year-over-year comparisons and to arrive at a calculation of what the owners of the company are actually getting out of it (and therefore what the shorts would have to cough up).

The first necessary adjustment is often to separate a company’s core operations from their non-operating income and expenses. In Red Hat’s case these adjustment are significant, because Red Hat has a vast portfolio of cash and securities that composes the bulk of its balance sheet. It also has currency translation issues as the firm does business in multiple regions. As interest rates are lower this year than last, it would be reasonable to eliminate interest, gains on trading in securities, and currency translation, to make the two periods comparable.

Now, it may be objected that I want to take out interest and trading profits because I’m approaching this situation from the short side and as a result I want to make the figures look as bad as possible. However, I would reply that even if the company may be entitled to a high multiple, their cash and investable securities are not. Their portfolio certainly doesn’t become magical because they own it. So, if we separate the $1.05 billion in cash and securities on Red Hat’s balance sheet, and treat it separately from the remaining $6.66 billion price, we get closer to looking at the value of Red Hat’s core operations.

This means lowering the 2nd quarter fiscal year 2011 earnings by $2.3 million, and the 2010 earnings by 5.7 million, to reflect interest, trading, and currency translation income. I should also note that 2010’s 2nd quarter tax rates were unrealistically low and should be adjusted. Taking $2.3 million out of pretax earnings and applying a standard 35% tax rate, 2011’s 2nd quarter “core” earnings were $22.2 million, and after a similar adjustment to 2010’s earnings we see “core” earnings of $17.9 million. Adding back in the difference between depreciation and expenditure calculated above, we get “core” free cash flow estimates of $24.2 million for 2nd quarter 2011, and $21.6 million for 2nd quarter 2010. So, at the end of this laborious process we have found that their core operational earnings growth is 12% over last year.

If we apply a similar adjustment process (eliminating interest and other income, applying a standard 35% tax rate, and adding back in depreciation but removing capital expenditures) across the last three years’ results, we find 2010’s full year “core” results were $81.8 million, 2009’s were $62.4 million, and 2008’s were $27.3 million (although capital expenditures that year seem unusually high).  So the growth is there, but it’s slowing (12% as compared to the year-ago quarter, down from 31.1% 2010 as compared to 2009 and 128% as compared to 2008, which, again, may be distorted by capital expenditures).

Turning now to the deferred revenue issue, Red Hat produces cash flow greater than its earnings because it sells year-long or multi-year subscriptions. The firm recognizes the revenue over the lifetime of the contract but on the whole likes to be paid a significant amount up-front. The cash that the subscribers pay in advance shows up on the cash flow statement and the balance sheet (offset by a liability for deferred revenue, of course), but it would be most unwise to consider it “free money.” Not only do those payments have to be earned by Red Hat providing the direct costs of the services that were subscribed, which costs money, but also they should bear their fair share of the cost of running the company, including administration, marketing and research and development. The deferred revenues should also bear their fair share of taxes, since the revenue most definitely will be taxed when it is recognized.

If we prorate these expenses between revenues recognized and deferred revenues, we can compute the effective operating margins for their core operations, and if we then apply these margins to deferred revenues we will be pretty close to calculating the company’s core free cash flow, which is what is needed to perform a sensible valuation in the first place.

In fiscal year 2008, the core after-tax operating margin including deferred revenues was 16.4% and accordingly we add 16.4% of their deferred revenues to their 2008 core earnings, producing a total estimated core free cash flow from operations of $42.2 million. In fiscal year 2009 the margin was 17.2% and adding back in that amount of deferred revenues gives us estimated core free cash flow of $79.3 million, and in fiscal year 2010 the margin was 16.5%, so adding back in that proportion of deferred revenue we get $95.4 million. This gives us growth rates of 73% between 2008 and 2009, and 20.4% between 2009 and 2010. Performing a similar calculation for 2nd quarter of 2011, we get core margins of 14%, giving us core free cash flow of $26.1 million, as compared to a margin of 14.5% and core free cash flow of $23.1 for 2nd quarter 2010. This gives us 13% growth, in line with what we calculated above. Fortunately the margins are stable enough not to produce a major distortionate effect.

I will also note that deferred revenues have been declining on the statement of cash flows since fiscal year 2008, which is indicative of deceleration of the growth rate of subscriptions.

Now, growth rates comparing one period to another give us the situation relative to how it was, but if we want to decide if a company is ultimately overvalued or not we have to use absolute measures, and our calculation of core free cash flow serves us well here. We already have the figure for the 2nd quarter of 2011, and calculating it for the prior  three quarters we have current core free cash flow of $99.8 million. When set against the $6.66 billion price the market is putting on Red Hat’s core operations, we have a P/E ratio of 66.7. So it is clear that the market is pricing in a high level of future growth.

How high, exactly? Well, let us use the H-model I have previously discussed. The H Model assumes that a company will grow at a high rate that will decay in a linear fashion to a stable growth rate (most companies show growth decay in a faster than linear fashion, actually). In The Dark Side of Valuation, Damodaran calculates that even starting from the IPO, median growth levels decay exponentially until by year 6 the company’s growth rate is indistinguishable from the other companies in its sector. So, six years is a good starting assumption.

The H model, in case you were wondering, is price = e*(1 + g)/(K-g) +e*H(G – g)/(K – g), where

e = current earnings
g = long-term stable growth rate
G = initial high growth rate
K = required return on equity investments
H = half the number of years until stable growth rate is reached.

Let us make some optimistic assumptions about Red Hat’s future growth. The initial growth rate is set at 30%, much higher than it is now. If 6% is the long term growth rate of Red Hat’s target market (generous but not unreasonable, as it takes into account both inflation and actual growth), and using the six year assumption and a required return of 10%, we get a total value for the core cash flows of the firm of $4.4 billion, far below the $6.66 billion price tag. In order for the company to grow into its valuation, then, either the initial growth rate has to be set at around 60%, or the high growth period has to last for 13 years instead of 6, neither of which I find particularly likely considering that growth is already showing signs of slowing.

I do not know what the reason for the optimism regarding Red Hat Inc.’s share price; there is always surely some residual love of technology at work in the investment community, and also perhaps people are looking at the gross margin of the subscription business, which is over 93% last quarter. However, the subscription business cannot be capable of limitless growth, particularly without the marketing, administrative, and research and development expenses necessary to support an expansion. Therefore, I find it highly unlikely that this company can produce the kind of growth necessary to justify its valuation. And so, although the company is operationally stable and does not seem to have any catalyst that would instigate a price drop, I can see ample grounds for considering it a short candidate.

Montier, in his Value Investing, made a spirited defense of value investing by demonstrating, not only that it works in a normal market, but also, based on the history of Japan, that it works effectively in a lost decade scenario. I find this significant because it strikes me that the United States faces a significant probability of facing such a scenario. Sadly, Montier couldn’t prove that value investing worked in a Great Depression scenario, which would make me feel better. But at any rate, a value strategy, he calculated, produced 3% annual returns in a period with very low inflation, when buy and hold returned -4% over the same period. But a long value/short strategy beat the pants off of both of them: 12%.

His book also contained advice about how to find a good short candidate. He expressed skepticism in the price/sales ratio, which he considers a transparent attempt to move up the income statement until something looks good. Perhaps the best that can be said about price/sales ratio is that it ignores both profit margins and capital structure. For example, Bon-Ton department stores has a price/sales ratio of 0.05, but since virtually all of its earnings are eaten up by interest, the stock is still highly unattractive.

But even if a low price/sales ratio does not necessarily indicate an attractive buy, an astronomically high price/sales ratio might well indicate an attractive short. Montier’s book includes a quote from Sun Microsystem’s own CEO to the effect that his firm was too expensive at a price/sales ratio of 10. The CEO stated that this lofty valuation assumes that his company purchases no raw materials or capital, that his staff works for free, and that despite his company not buying anything or paying anyone, it pays no taxes.

In this vein, I am pleased to present Concur Technologies (CNQR) as a short selling candidate. I hope I shan’t be accused of piling on, but the apparent level of optimism around this stock seems to have surpassed unrealistic and moved on to fanatical. The company has a price/sales ratio of 9.4 and a P/E ratio of 117. True, they have produced five year earnings growth of 33% per this article at Motley Fool, but the question then becomes how many years they can sustain that level of earnings growth, and how many years they have to sustain it in order to justify their current valuation.

Fortunately, Damodaran, in Damodaran on Valuation, has given us a formula for evaluating companies that are currently growing at faster than stable rates. The H Model described in his book assumes that a company that is now growing at a high rate will see its growth rate decay in a linear fashion to a stable growth rate. It may be objected that most growth companies do not actually show a linear decay in earnings growth rate, and I must say that I agree with that sentiment: growth usually decays in a faster than linear manner. In The Dark Side of Valuation, Damodaran calculates that even starting from the IPO, median growth levels decay exponentially until by year 6 the company’s growth rate is indistinguishable from the other companies in its sector.

The H model, in case you were wondering, is price = e*(1 + g)/(K-g) +e*H(G – g)/(K – g), where

e = current earnings
g = long-term stable growth rate
G = initial high growth rate
K = required return on equity investments
H = half the number of years until stable growth rate is reached.

It’s not the most sophisticated formula, but it does give you a rough idea of what the market is expecting, and is not calculation-intensive enough to be inconvenient.

So, Concur’s current earnings per share are 44 cents, but they did record a significant noncash interest expense of $2.5 million relating to their convertible debt, so if we put that back in the earnings improve to 49 cents. Let us further assume that the high growth rate of 33% holds true, and that the long-term growth rate is 6%. This is not unreasonable since it includes inflation as well as economic growth. It is optimistic, but when we consider short candidates we want to make optimistic assumptions, just as when we want to value longs we want to make conservative assumptions.

Assume further that the high growth period will last six more years, and that we have a required return on equity investments of 10%. This gives us $22.91, which is less than half of the current share price.

Now, as I said, the H model gives us an idea of what the market is expecting, so let us solve for what the market actually is expecting. If the market accepts 33% as the high growth rate, for example, how long is the market expecting the company to go before it drops to stable growth? Using the H formula, we find that the market expects the high growth period to last 23 years, which, considering that 6 or less is more common, strikes me as somewhat unlikely. It assumes that even nine years from now they will still be producing 20% growth.

Or, we could assume that the six years is correct, but that the company is entitled to a higher high growth rate. Keeping everything else constant and using a six year assumption, we find that the market is expecting earnings growth rates of 109%, meaning that the firm will double its earnings growth next year, nearly double them again the year after that, make a further increase of almost 3/4 the year thereafter, and so forth. I find this outcome also highly unlikely.

If we take a hybrid assumption of 50% for the high growth rate and 10 years of excess growth, we still get only $39.44, about 23% below the current share price. So, even results that most companies would kill for would not be good enough to justify buying Concur at its current price.

But perhaps there is something about Concur that makes it so interesting. Its primary line of business is automating tracking and reimbursement of employee spending on matters such as business travel. Increased automation of these processes decreases both time expended and error rates, as it tends to cut down on businesses using the manila-envelope-full-of-receipts method of filing. However, it does not appear to me that they have any sort of defense from the competition; expense management software is not so unique, and even if their software-as-a-service model allows them more flexible pricing given the needs of their customers, their competition could take advantage of the same service.

Furthermore, and perhaps more disturbing given the hyper-growth thesis that the market currently holds for Concur, their growth seems to have hit a ceiling.  Between 2007 and 2008, sales nearly doubled and profits more than doubled. Between 2008 and 2009, sales increased by over 15%, from $206 million to $239 million, and net income increased by a little under 50%, from $17.2 million to $25.7 million. However, 2010 year to date, sales have increased an additional 18%, and yet income to date, even putting back the noncash interest expense, has in fact declined slightly as compared to last year.

Furthermore, one aspect of the company that gives me further concern on the long side is that more than half of the company’s assets, $550 million out of $950 million total, consists of cash and short term investments. It seems to me that a bona fide growth company should be able to find a better use for its money than just sitting on it. The company’s return on non-cash assets is a robust but not too outsized 7%, so if all their cash was in fact converted into usable assets earnings would theoretically more than double. So, the fact that this has not occurred would only be because the firm has not yet found the customers to allow it. And even if it did occur, this trick would double earnings once but not twice.

As a final aside, I will note that the firm has outstanding $285 million in outstanding convertible bonds, an additional 5.4 million shares, or roughly 10% of total shares outstanding, that are bumping against the conversion price. Curiously, the firm purchased roughly 5 million shares worth of call options on itself to offset the effect of the bonds converting, and the $60 million in options premiums sort of defeats the purpose of saving money by issuing convertible debt, or so it seems to me.

The only thing missing from making Concur into the perfect short is a catalyst. The company’s interest requirements are covered by about nine times, so there is no apparent risk of distress. The only real problem with this company is the price, and, sadly for shorts, overvaluation can persist for some time. However, given the massive levels of growth required to make the current share price have any semblance of logic, it strikes me as unlikely that the company can pull it off, particularly if competition starts sniffing around. And so, I would conclude that Concur Technology is an attractive short candidate.

Now that the BP well has been finally containmed, it is only natural that market participants have come to expect a great deal of the shrinkage in valuation of Gulf oil producers to reverse itself to some degree. Certainly there will be new regulations, and perhaps more onerous inspection standards, but on the whole it seems that people are expecting some return to normalcy in Gulf extraction operations.

However, even if it is likely that many Gulf oil extractors will rebound up to, or even above, the price levels prevailing before April 20th, not all of them deserve that rebound. Some of them were in fact overpriced to begin with.

Stone Energy (SGY) is an example: on April 20th, the day the oil rig exploded, their stock closed at 18.42.  It dropped to a low of 10.30, and as of September 13th, closed at 12.44. Normally one would consider this company as primed for considerable appreciation now that the oil spill is contained and the Gulf is attempting to move on.

However, a closer look at Stone Energy’s free cash flow suggests otherwise. Stone Energy currently trades at a reasonable, if not robust, P/E ratio of 14.18, but the free cash flow picture tells a different story.

Free cash flow is the amount of a company’s earnings that it can actually distribute without affecting its earnings power. The notion of free cash flow as a thing separate from accounting earnings goes back to Benjamin Graham, who reminds us that the “true” profits of a business can only be calculated after deducting whatever expenditures allow the business to maintain its earnings power. Sometimes depreciation is a proxy for these expenditures, but in many cases, capital expenditures on a forward-looking basis have nothing to do with a backwards-looking measure like depreciation. As a result, a convenient estimate for free cash flow for a given period is earnings plus depreciation minus capital expenditures.

I am aware that not all capital expenditures are intended to replace used-up earnings power, and that for many companies, some or most of capital expenditures are intended to expand earnings power rather than simply maintain it. However, no company ever differentiates maintenance capital expenditures from growth expenditures in its financial reports. However, the conservatism necessary for investing in general, and especially for oil extraction companies, which by their nature invest in assets with finite lives and thus are constantly in need of replacement assets, it is advisable to treat all capital expenditures as maintenance expenditures unless there is a very good reason not to.

Stone Energy’s projected capital budget in 2010 is $400 million, of which $170 million has been spent to date. In 2009, the figure was $320 million, in 2008 the figure came to $450 million, and in 2007, 230 million, not counting an acquisition in 2008 that was paid for mainly in stock. So it would seem that the figure of $400 million is not of an outsize proportion that could be dismissed as nonrecurring.

Year to date, Stone Energy has reported earnings of $56 million and depreciation of $124 million, not appreciably better than results of the first two quarters last year (not counting a nonrecurring impairment that I shall have more to speak of later). Taking out the $170 million in capital expenditures year to date, this gives us free cash flow year to date of zero. This represents an improvement over last year’s estimated free cash flow of minus $36 million for the first two quarters, but still nothing that would justify a great deal of optimism, particularly with capital investments estimated to come in $60 million higher over the latter half of the year.

For the full year 2009, free cash flow was $60 million (neglecting the large impairment and the tax writeoffs that it produced). In 2008, estimated free cash flow as adjusted was $175 million, and in 2007, free cash flow was $252 million.

Much of this diminution in earnings is the result of the dramatic fall in oil prices from their 2008 peak, which was also the cause of a writeoff of $466 million in goodwill and $1.814 billion in property values. Although the impairments are themselves a noncash expense, it is clear that their effects on earnings power have not been exaggerated.

So, even though Stone Energy may see a recovery in its share prices as a result of a broad recovery in Gulf oil produdcers’ prices, it does not appear that the current free cash flow levels would support it. It would be more prudent to look elsewhere for a more sustainable recovery from the BP-caused lows.

Since my last foray into the taxation arena, where I examined the (most probably nonexistent) link between tax collections and growth, proved so interesting, even though it was a purely statistical analysis and nothing to do with tax policy as such, I thought I would make a modest proposal in some actual tax policy directions. As I have had occasion to state in the past, I’m not a macroeconomist, but that never stops anyone else from discussing macroeconomic issues, and it won’t stop me. I have two ideas, one “modest,” and one real.

It has been estimated that every dollar cut from the IRS’s enforcement budget winds up costing the government four dollars in revenues. And so it seems to me that if the IRS’s auditing staff were expanded by 10 or 20 or 40 thousand people, or even more, we could simultaneously lower unemployment and cut the deficit. As such, it is a win-win situation–apart from the fact that it would possibly be the least popular thing this administration could do.

However, one idea that I think is more promising is the elimination of the tax deduction for amortization of goodwill. As you know, goodwill is created when a firm pays more than the value of the assets in an acquisition; it is the difference between the purchase price and the asset value. For many firms, particularly serial acquirers, this can add up to a significant amount. Under current tax law, goodwill amortizes over 15 years.

I suppose the purpose originally was to treat intangible assets the same way as tangible assets, which also depreciate over time. However, as I have had occasion to state before, goodwill is not a “real” asset the way intellectual property or trademarks or brand name recognition are real assets. It cannot be licensed, sold, nor its value otherwise tapped, apart from through the tax deduction itself. Many value investors advocate ignoring goodwill, both on the balance sheet and when it is impaired on the income statement. Our reasons for doing so are quite simple: goodwill represents the theoretical excess earnings power arising from acquiring a company rather than just replicating its assets. If that earning power persists, it will be reflected in the calculation of income and return on assets, and if that earnings power drops, it will also be reflected. Therefore, we do not require goodwill or its amortization to tell us what we already know.

The trouble with amortization of goodwill is that it effectively results in taxpayers subsidizing a merger. This paper found a statistically significant increase in goodwill paid for acquisitions after the tax laws changed to allow goodwill amortization.  Whatever amount of goodwill is created by a merger, the taxable income of the merged corporation will be reduced by 1/15th of that amount every year for the next 15 years, with the rest of us taxpayers picking up the tab. The corporate tax rate in the United States is currently 35%, and considering the massive amounts of goodwill bouncing around the system, the total amount of amortization deductions easily reaches into the tens of billions per year.

Of particular concern is when two firms get into a bidding war over an acquisition target, where prices can be bid up to giddy heights far removed from tangible valuation. In contentious auctions, it is often the case that the winner of the auction is actually the one who comes in second, because he or she has convinced a competitor, the actual winner, to pay too much. However, as the price being bid up generally results in an increase in the goodwill portion of the total price paid for the acquisition, the taxpayer is in effect subsidizing fully 35% of the bid increases.

Furthermore, let us consider that mergers and acquisitions have a history that is mixed at best. Damodaran, in his Investment Fables, tells us that researchers examined a set of mergers between 1972 and 1983, with an eye to two questions: 1, Did the return on the amount invested in the acquisition exceed the acquirer’s cost of capital, and 2, Did the acquisitions help the acquirer outperform the competition. 28 of 58 mergers sampled failed both tests, and 6 more failed at least one. Furthermore, many acquisitions are reversed within short time periods; one study places the figure at 44% of the total with a short lead time, and cited that the most common grounds for reversal were that the acquirer overpaid or that the operations of the firms did not mesh. Over longer periods the divestiture rate has approached 50%. We can all understand the virtue in subsidizing success–if mergers are successful a firm’s taxable income will increase, with the result that the deductible amortization  may be offset in part, in whole, or even in excess, assuming that the additional profits could only have been unlocked by the merger itself. But, if mergers have high odds of not producing the anticipated synergy, taxpayers are essentially subsidizing failure.

This is offensive enough in a normal environment, but it is particularly troublesome in the current environment, where firms are accused of piling up assets and either engaging in or preparing for mergers rather than spending that money on internal investment and hiring. Banks fell under considerable criticism for using their TARP money to acquire weaker competitors back in 2008 and 2009.

In summary, the elimination of the amortization of goodwill would remove the incentive for companies to invest their resources in ill-starred and overpriced mergers, and allow them to focus on building better companies, not merely bigger companies. It would also provide significant deficit reduction at a time when the nation is in great need of it.

I am recommending stock in Entercom Communications Corp (ETM), on the grounds that they are considerably undervalued. Of course, I wish I had issued this recommendation last week when they closed at $5.29 per share, but even now that they are at $7.30 per share, the company is still compellingly undervalued.

Entercom claims to be “one of the five largest radio broadcasting companies” in the United States, which I take to mean that they are the fifth largest. Like so many companies, they recognized large impairments of their intangible assets in 2008 and 2009, which made earnings look absolutely horrible for those two years. However, such impairments are a noncash expense, and based on the firm’s actual free cash flow, Entercom produced, and continues to produce, a very attractive return based on the current share price.

In 2008, for example, the firm booked an operating loss of $710 million. However, $835 million of that loss was due to a large impairment; without that, the firm would have produced operating income of $125 million. After removing $45 million in interest, and applying a 35% tax rate, we get $52 million in sustainable earnings. For 2009, when advertising revenue had dropped somewhat from 2008, the firm reported operating income of $11 million, but that was after an additional $68 million noncash impairment, and as adjusted the firm produced $79 million in operating income. After removing $31 million in interest (much of Entercom’s financing is from variable rate debt) and applying a 35% tax rate, we get $31 million in sustainable earnings. For the first two quarters of 2010, advertising revenues have rebounded a bit as compared to the first half of 2009, producing operating income of $41 million, and no impairments at all, finally. After $14 million in interest, and applying a 35% tax rate again, we get earnings on a sustainable basis of $17.5 million, or $35 million projected for the entire year. Since the company at the current price has a market cap of $260 million, this produces a P/E ratio of 7.5.

If, however, we use our site’s favorite proxy for free cash flow, the situation improves further, since in 2008 Entercom took a depreciation and amortization allowance of $20.5 million, but made capital expenditures of only $8.5 million. In 2009, their depreciation allowance was $16.5 million, but they made only $2.5 million in capital expenditures. And in 2010 year to date, depreciation ran $6.5 million, but capital expenditures have run only $1.5 million. So, it would seem, if current levels of capital expenditure are deemed adequate, that the firm has an additional $10 million in free cash flows at least, which would give it projected 2010 annual free cash flow of $45 million, and a price/free cash flow ratio of 5.75.

Furthermore, Entercom, according to its latest 10-K, had a net operating loss carryforward of $48 million as of December 2009, which should serve to shelter their income from taxes for all of 2010 and possibly part of 2011 as well. This would save the company $16.8 million dollars.

I will admit that Entercom’s balance sheet is looking more than a little anemic; the stated value of their assets, owing to the massive impairments, is $913 million, and they have $777 million in total liabilities, producing shareholders’ equity of $136 million. However, I believe that these impairments were overstated, because I find it hard to believe that $136 million in assets (actually, 114 million at the end of year 2009), can produce free cash flows of $45 million; that’s a return on equity of roughly 33 percent which strikes me as unusually high. Furthermore, in terms of interest coverage, the company seems to be doing adequately: 2.75x in 2008, 2.5x in 2009, and 2.9x year to date 2010. This figure approaches the 3x coverage that I use as a rule of thumb to allow debt to appear reasonably safe. As I stated, though, much of Entercom’s debts are variable rate. On the plus side, Entercom has interest rate derivatives to prevent them from rising interest rates in the notional amount of $475 million, roughly 2/3 of their total long term debt. On the plus side, Entercom has been paying down their debt at a fairly rapid pace, $76 million in 2009 and $35 million in 2010 to date. And, realistically, given the current environment I don’t see interest rates rising by much any time soon.

So, despite the 38% run-up in a week that I missed (I’m sorry), I think that Entercom is still dramatically undervalued.

As I seem to be in a mood to find junk bonds lately, I am pleased to present you with the bonds of the Mohegan Tribal Gaming Authority. The Mohegan Tribe operates a massive casino in Connecticut, one of only two casinos in New England, as well as several other ventures.

The company has been attempting to open casinos on other Indian reservations in the recent few years, but owing to certain regulatory difficulties, both proposed casinos have been put on hold. Coupled with the drop in revenues from the economic slowdown, which seem to be leveling out, the ability of the Mohegan Authority to service its debts has come into question. Although their operating incomes have never actually failed to cover their interest requirements of roughly $100 million per year, they certainly have come close, as I shall state below.

The Mohegan Authority’s financial statements require two species of adjustment. The first is the one I customarily make, which is adding back depreciation and removing capital expenditures in order to reach a convenient proxy for free cash flow to the firm. In Mohegan’s case, though, the numerous projects they embarked on, including the drive to open two other casinos, a major expansion to their existing casino (now on hold), and the purchase of a WNBA franchise, has served to give their capital investments a remarkably uneven quality from year to year. As a result, I can find no better course than to average out their capital expenditures for a considerable number of years, which comes to about $70 million in capital spending per year. It may be that given the apparent regulatory opposition to expanding casino gambling as well as the economic slowdown, the company may be in a position to ratchet down their future capital expenditures. This would increase free cash flow, but with so many projects on hold indefinitely, we may expect to see (noncash) writeoffs in future as well, which may drag on the price of the bonds.

The financial statements also require another adjustment, and this one is company-specific. The Mohegan Authority made a deal with a developer in 1998 that requires them to hand over 5% of their revenues until 2014. The company records their projected payments under this deal as a liability. In my view, this figure should not be treated as a liability, since the only way for the liability to accrue is for the company to do business and so it should simply be treated as an operating expense. In fact, the current treatment leads to the paradoxical effect that the company’s liability goes up the better the company does. Accordingly, when calculating operating income to arrive at interest coverage, the projected change in this liability, which is separate from amounts actually paid under the agreement, should also be stripped out.

So, adding back in the change in the liability (which can be negative, as it was in 2008 and 2009), adding back in depreciation, and removing $70 million from operating income to account for capital expenditures (or a prorated amount for the 3 quarters of fiscal year 2010), we get the following figures for available operating income:

2004: 180
2005: 192
2006: 208
2007: 232
2008: 123
2009: 126
2010 ytd: 159

I would like to focus on the last three years, because two of them were very low years, and the current year may give us some insight into where the company stands now. The Mohegan Authority has several classes of long-term debt, including $527 million in bank loans, $193 million in privately placed 11.5% senior notes, $250 million of 6.125% senior notes due 2013, and a family of subordinated notes with $475 million outstanding. Adding back in payables and miscellaneous liabilities and removing the liability for their deal with the developer, the Mohegan authority has total debt of approximately $1.9 billion, set against $2.27 billion in assets, most of them long-term and/or intangible.

In terms of interest coverage, the total interest paid year to date, apart from amortizations and other noncash items, comes to $82 million so far this year. This represents an interest coverage ratio of nearly 2x, and typically the summer is prime gambling season at the Mohegan Authority, so the full year results may show even more improvement. However, in 2008 and 2009, interest paid was $96 million and $102.4 million, resulting in interest coverage ratios of 1.28x and 1.24x. However, the senior bond, has a coverage ratio of 3.5x in even the worst years when one strips out the interest of all bonds subordinate to itself.

So the question becomes which class of debt to buy; the 6.125% senior notes or one of the subordinated notes. Both classes of debt are rated CCC+, which seems to be a favorite rating around here; Bon-Ton’s bonds and WNR’s convertible bonds are also CCC+, and it may be that the ratings agencies reserve this rating for bonds that could conceivably justify a rating in the B’s, but don’t want to go out on a limb by actually giving them one. Especially now that the subprime debacle has naturally made them worried about handing out ratings that are too high.

The 6.125% senior notes due 2013 currently trade at a 79.50, has a current yield of 7.7% and a yield to maturity of 16.6%, which would be a respectable return for anyone. Among the three issues of subordinate debt, the 8% notes due 2012 trade at 79, currently yield 10.25% and offer a yield to maturity of 25.8%; the 7.125% notes due 2014 trade at 60, currently yield 12.18% and offer a yield to maturity of 23.9%, and the 6.875% notes due 2015 trade at about 55, offer a current yield of 12 and a yield to maturity of 24% (my bond screener says that there is a bid/ask spread of 8 on these bonds, so I’m just averaging the two).

I am of two minds when it comes to the relative safety of the senior versus subordinate bonds. Benjamin Graham, in his excellent Security Analysis , advocated satisfying oneself that all of the bonds of a company are safe and then buying the highest yielding one, but that was his advice for generally investment-grade debt. On the other hand, as stated in Stephen Moyer’s vitally useful guide to distressed and bankruptcy investing, Distressed Debt Analysis, seniority really comes into play in a bankruptcy. However, I am not sure how a bankruptcy would affect an Indian casino, because under the law only a tribe may own an interest in an Indian casino, so the usual outcome of a class of bondholders becoming the new stockholders is unavailable. Even so, any workout would result in the senior debtholders taking much less of a haircut.

That said, with the lower capital requirements from stalled expansion projects, I don’t think that bankruptcy is anywhere near imminent, although the Mohegan Authority does seem to have to roll over their debts at higher yields. So the 8% subordinate bond due April 2012 would seem attractive at first glance, with a yield of 25% and a short time horizon (only 19 months). But if you find yourself asking “What could possibly happen in 19 months?”, ask yourself what happened in the 19 months starting in April of 2007. Given the currently tenuous situation, I do think the more defensive approach of the 6.125% senior notes, with their current yield of 7.7% and a yield to maturity of 16.6%, offer an excellent return. One can always move down the seniority ladder when and if the situation at the Mohegan Tribal Gaming Authority improves.

PS: I am aware that certain followers of socially responsible investing would prefer not to receive income from activities such as gambling. I take the position that since investments purchased in the secondary market do not result in any cash actually going to the company in question, then receiving the income from the bonds does not equate to supporting gambling. In fact, there would be a certain poetic irony if, say, Gamblers Anonymous invested its spare cash in the securities of casino operators so it could direct the casinos’ own profits against them (or from a more cynical standpoint, it’s hedging its bets). Or, if I may coin the expression, It is not what goeth into a man’s pockets that defiles him, but what cometh out of a man’s pockets that defiles him.

I noticed on Yahoo! finance a couple of days ago that Paul Otellini, CEO of Intel, publicly complained that, owing to high taxes and regulatory uncertainty, innovation and growth in the US would be stifled and, of course, Obama is to blame.

Of course, this would be the same Intel that paid AMD a $1.25 billion settlement over antitrust violations, settled claim with the FTC a couple of weeks ago over its anticompetitive activities, and is still fighting off a $1.45 billion claim in Europe. Of course, antitrust laws have been on the books for more than a century (and were invented by Republicans), so if the CEO hasn’t quite got the hang of complying with them after all this time, then his complaints come off as a case of sour grapes to me.

He did also state that we have to look forward to “an inevitable erosion and shift of wealth.” Now, it is only natural for a business to want less regulation, and of course it is natural for everyone to want to pay less taxes. But these comments got me thinking of how strong the relationship is between taxation and growth, particularly now that the Bush tax cuts are due to expire.

The job of an economist, it has been said, is to take something that works in practice and make it work in theory. The author of that statement may have intended it as a satire, but it is actually a remarkably accurate description. And in theory, too high a level of taxation on production forces economic participants to direct more and more of the nontaxed remainder of their economic activity towards immediate consumption rather than the building up of capital, and also makes certain marginal uses of capital economically unattractive, causing its owners to leave it idle or consume it. And without capital there can be no growth.

So, that is the hypothesis that economists and CEOs have set up for us: more taxes = less growth. Fortunately, we have a way to test this hypothesis, as the government, thanks in part to the taxes it collects, has given us a wealth of data on historical GDP levels, and also of taxes collected as a portion of GDP. With these two pieces of data, we can test the hypothesis.

If, as it is stated, more taxes = less growth, then we should see lower growth in a period where taxes as a portion of GDP are high, and higher growth in a period where taxes as a portion of GDP are low, and using the elementary statistical analysis technique of regression, we can calculate the r-squared, the coefficient of determination, which will tell us how accurately the data fit the theory. An r-squared close to 1 means that there is perfect correlation, and an r-squared close to 0 means that there is no correlation.

The most basic mode of analysis is to compare the taxes collected in a given year with GDP growth that year. Using a linear regression function and taking data since 1963 (a year I chose more or less arbitrarily) up until 2007, it gave me the formula of y = 13.73 – .56x, , where y is GDP growth and x is the level of taxation as a portion of GDP. Now at first glance the hypothesis is correct; every 1% increase in taxation as a percentage of GDP results in GDP growth that year dropping by .56%. However, would you care to know what the r-squared was? .07. Remember how I said that an r-squared close to 0 suggest complete randomness? So in theory, tax levels explain 7% of GDP growth (the other 93% are explained by something else), but with an r-squared that low, I am forced to question the validity of the expression itself.

But since I have the data I may as well play with it. After all, very often one only knows one’s tax liability after it’s too late to change it, and one just has to adjust in future periods accordingly. So it would perhaps be more fair to look at the taxes collected in one year and the effect on growth for the next year. Here I get a more dramatic looking equation, y = 16.49 -.72x, but again my coefficient of determination is only .11, meaning, depending on interpretation, that my equation has an 11% explanatory power, or that there’s only an 11% chance that I’ve discovered anything at all.

But while I’m at it, I recall that capital and consumption decisions are often made with more than a year of time horizon. From that perspective, it would probably be more enlightening to compare taxes collected in one year with GDP growth over the next three years. This time I get y = 6.59 -.19x, and my r-squared is only .02. So in fact what logic suggests would be a clearer relationship is actually the more tenuous one, as the r-squared suggests that the odds are very good that the relationship is nonexistent.

And finally, I am reminded that capital is a thing that accumulates over time based on previous consumption and savings decisions, so in the interest of thoroughness as well as symmetry, I compared average taxation as a percentage of GDP over the last three years to average GDP growth over the next three years, and I was amazed at what I found. y = 4.34 + .09x, indicating that paradoxically, higher tax collections over a three-year period actually lead to higher growth over the next three years. It made no sense to me, until I consulted the r-squared figure, which was .003, meaning pretty much conclusively that there was no detectable correlation at all.

And just to finish out the available set of comparisons, calculating the effect of three-year average tax collections on growth in the subsequent year produces an r-squared of .016, no better than most of the other experiments and also far below a level of significance necessary to conclude that a correlation exists.

So, the upshot of this fairly basic analysis is that no matter how you slice the data, it seems that tax levels (at least the tax levels that have prevailed in the US since 1963) have an undetectable effect on GDP growth, if indeed they have any at all, and that therefore GDP growth must come from other factors.

I am aware that since the data for taxes as a portion of GDP has only one decimal, it is entirely possible that I have two GDP growth figures for a single level of tax, so this method could be criticized that if one line has to go through two points on the same point on the x axis, it is inevitable that it can only be close to one of them. However, I would counter these critics by asking why, if tax levels have a strong explanatory power, would the two growth points be far away from each other in the first place.

It may further be objected that taxes and GDP growth are a chaotic and nonlinear system, and so the use of a linear regression would not be reasonably calculated to produce a good fit. But a cursory visual examination of the raw data indicates that there is no obvious cluster, and using the other regression tools available in Excel produce no improvement.

So, it seems, based on this simple analysis, that cutting taxes as a portion of GDP has a poor record in producing near-term growth, and so the perennial push to make further tax cuts in order to juice economic growth would appear misguided. But on the plus side, the impending expiration of the Bush tax cut is equally unlikely to strangle an economic recovery, as many people have feared.

I find that the stocks I like seem to come in pairs. I liked oil and gas producers Linn Energy and Breitburn; I liked phone companies Windstream and Qwest, and now in the contract manufacturing sectors, I liked Keytronic and now I like Amkor Technology (AMKR).

Amkor produces semiconductor packaging for other manufacturers, and also performs testing services. Most of their operations are in Asia. Based on today’s prices, I would estimate their P/E ratio at a little over 7. I shall go into more detail on that later.

On the downside, Amkor has significant debt; $2.7 billion in total assets and $2.2 billion in liabilities, producing a price/book ratio of around 2. Although they have significant R & D expenditures, and, following Damodaran, in Damodaran on Valuation, I have advocated capitalizing these, their interest coverage is still hovering around a little over 2x, firmly in junk status, although their credit rating was recently raised a notch to BB-.

Furthermore, some of that debt is convertible at various prices, ranging from $14.59 per share to $250 million in notes at $3.02 per share (the price as of Monday’s close is $5.61). As a result, there is a significant gap between normal and fully diluted earnings. The $3.02 convertible notes do deserve further attention; they were sold to the company’s chairman in April 2009, and the conversion price is actually lower than the price of the stock was at any time during the month. Most convertible notes are issued with a conversion price that is at a significant premium to the current stock price. Now, I know that April 2009 was still a time of economic crisis and was only a month after the market hit bottom, and as a result financing was hard to come by, but even so this level of blatant self-dealing should be a black mark against the firm.

Amkor is a cyclical company, and as a result sales for 2009 were no higher than they were in 2005. However, apart from a goodwill writeoff in 2008, the firm has been profitable every year since 2006. The recent improvement in the semiconductor device market has given the firm some improvement in their results as compared to last year. Turning now to earnings, in 2009 they reported earnings of $156 million, but because of operating losses, instead of paying taxes it worked out that the government actually owed them. If their tax liability had been normal, they would have earned only $80 million that year. They took $300 million in depreciation and amortization, and made $235 million in investments, producing free cash flow of $155 million. In 2008 the figure was $275 million, and in 2007, about $220 million.

Year to date 2010, they have reported earnings of $104 million, which is again propped up by tax effects so the correct figure would be about $67 million. Depreciation year to date has been $154 million, and capital expenditures $143 million, producing free cash flow of about $78 million, or $156 million on a full year basis. The firm also disclosed in their SEC filings that they typically see better results in the last two quarters of the year, and also that they intended to front-load a significant amount of capital expenditures for the year, so it is possible that the full year’s results will be even better. Furthermore, they have $300 million in net operating loss carryforwards to burn through as well.

I think the main drag on Amkor’s valuation is their significant debt levels, and although they have not made significant movement towards paying their debts down lately, which I would have preferred them to do, they have at least been refinancing at lower rates. During the last quarter, they borrowed what amounts to $133 million at a variable rate, currently 4.5%, to be paid down by 2013, to replace $125 million of 9 1/4% notes due 2016 (I assume some of the difference went to loan costs and the rest they kept). They also issued $345 million of 7 3/8% notes due 2018 (which they intend to replace with publicly traded notes by the end of the year, otherwise they will be liable for more interest), to replace their existing notes due 2011 and 2013. These transactions cost them $17.8 million, which is theoretically nonrecurring, but since the company has a large variety of debt outstanding and therefore has to engage in these operations frequently, and also since 2009’s results include a $16.8 million gain from similar transactions, I would not say that these costs are nonrecurring enough to eliminate from consideration. Pity they won’t give the notes convertible at $3.02 this kind of refinancing treatment.

As an aside, I will note that their 10-K is a thing of beauty. Very clear writing and a lot of easily digestible subheadings.

In the final analysis, I do think that the company is offering a very good return on investment and is capable of managing its debt load.

I have a certain weakness in my heart for junk bonds. Junk bonds, of course, are bonds with a sub-investment grade credit rating, and they can range from just barely on the cusp of investment grade to the point where the purchaser would have to be crazy to consider them. As Ben Graham wrote, an ordinary bond investor should never buy them, and speculators prefer a more speculative medium, and yet there are millions (now hundreds of billions or trillions) of them out there and someone has to own them.

And better still, some people can’t own them. Many institutions are limited to investment-grade holdings, so a cut in ratings means that they have to sell them. Any situation where investors refuse to buy a security for any reason other than its investment merits creates a fruitful hunting ground.

Of course, picking individual bonds is no easier than picking individual stocks, and given that the potential upside for a bond is limited to getting principal and interest while the potential upside for stock is theoretically unlimited, many people have called it less rewarding. However, the upside of bonds can be very large indeed if they sell at a discount, and the upside of stocks is effectively limited, at least for value investors, because a stock can only go up so far before it becomes too speculatively priced to consider. Even so, this limits us to bonds that are priced below par. A bond below par with a near maturity date is an advantageous situation, since the bond must be paid off at par or default, and as the time nears the uncertainty of eventual payoff lessens and such bonds creep towards par as if by magic–assuming that the company is capable of paying off the debt, typically by a new bond issuance.

Since a lot of such bonds are issued by marginal companies, it is only natural that these companies would use whatever methods they can to reduce their interest obligations. One of the most common methods is by issuing convertible bonds, whereby the bondholder can convert the bond into a given number of shares of stock that normally represents a price well above the market price. Basically, it acts like a long-lived call option.

Convertible bonds have been around for decades, and they in fact predate the options pricing models that let us figure out what they’re worth. Of course, a convertible trading at below par is probably trading at below par because the company that issued it has gone through some misfortunes since then, so the possibility of actually converting is not really much of an issue in the bonds’ valuation.

However, the accounting profession is never content unless it is making a simple thing complicated. For most classes of convertible bonds, accounting standards codification 470-20 requires them to divide the bonds between the debt on the balance sheet itself between the debt and the option component. This means that if a company borrows $300 million of convertible debt due in five years, and based on comparing interest rates, similar nonconvertible bonds would sell for $250 million, then the company would record a liability of $250 million, and every year take $10 million in additional interest costs to bring the liability back up to $300 million by the time the debt is due.

This is a useless complication that is completely inconsistent with everything most people know about accounting. It obfuscates the facts rather than makes them clear. The obligation to repay the extra $50 million doesn’t accrue in 5 years; it accrues immediately when the debt is issued (particularly since anyone familiar with options knows that it’s typically better to sell the option than exercise it, so anyone with a convertible bond that’s worth converting would probably rather write a call/buy a put and keep the bond, rather than actually convert it). The effect of this rule means that the company’s liabilities are constantly understated on the balance sheet, and that the firm has to record a noncash interest expense every year to amortize a discount that never existed in the first place.

I look upon this rule as a symptom of a disease that Marty Whitman diagnosed during the debate about expensing stock options: the assumption that financial statements should be geared only to equity holders. Stock options are an expense to equity holders because they result in dilution, but creditors don’t care if equity holders are diluted to nothing as long as they get paid. Here, too, a creditor looking at the fictitious financial statements required by 470-20 would conclude that a firm with convertible debt has much fewer liabilities and a much higher interest expense than the actual situation would suggest. Creditors other than the holders of the convertible bonds care primarily about interest coverage, and imposing a large noncash interest charge makes them feel unduly nervous. The holders of of the convertible bonds themselves couldn’t even figure out from the balance sheet what is the aggregate size of the bond issue they own. Only the equity holders would actually care about the dilution caused by converting.

As a final insult to reality, the allocation between debt and options is fixed when the bond is issued, so if the price of the company collapses the debt doesn’t automatically move closer to its full value despite the fact that conversion is highly unlikely. Without updating the figures in real time, the option premium contained in the convertible bonds tells no one reading the financial statement about the actual odds of dilution, which is what this rule was intended to accomplish.

I found out about this senseless rule while I was reading about Energy Control Devices’ bonds, which I ultimately decided against because, although the bonds are presently backed by substantial current assets, the firm itself has no operating income. Investing in such a firm basically involves placing a bet whether the firm is going to run out of money before the bonds fall due, and that sort of analysis is too reliant on foresight.

But I am considering the 2014 Western Refining (WNR) 5.75s, which sell at about 76, representing a yield to maturity of about 14% and a current yield of about 7.67%. The conversion price is around $10.80 per share, more than twice the current share price, so the possibility of conversion hardly seems to be an issue. The bonds’ credit rating is CCC+, which is the best of the bottom tier.

These convertible bonds are designated the senior bonds, but the other classes of bonds have security and guaranty provisions, so they are actually probably the junior debt. Western Refining’s total interest requirement year to date has been $61 million. Operating earnings year to date are $37 million, plus $69 million in depreciation, and $37 million in capital expenditures. Furthermore, the firm has listed $23 million in maintenance expenditures year to date. According to its financial statements, various classes of maintenance activities have a schedule of roughly 2-6 years between actions. Looking back at their history, we find that in the last five full years they have spent more than $23 million only once, and on average over that period maintenance expenditures were $16.5 million. If we assume that the company has spent an outsized amount on maintenance, and instead applied the average amount of maintenance expenditures over the years, we find that there is theoretically another $14 million in cash flows available to meet interest. This produces a theoretical operating cash flow of about $85 million to meet their interest requirements, and an interest coverage ratio of 1.25, which is consistent with a CCC or a CC rating.

Also, their working capital has expanded recently, resulting in a cash burn rate that would normally be disturbing for interest coverage issues. However, let us recall that Western Refining is an oil refinery, and the end of the last quarter coincided with the early phase of the summer driving season. Furthermore, the large amount of maintenance they did year to date has resulted in several weeks of a shutdown in production, which makes the year to date results unusually low.

Now, the fact that they are a refinery may also give investors some confidence; refining services will be in demand as long as oil is in use, and it may allow them to survive with lower interest rate coverage levels than a typical company that produces more discretionary products. However, refiners do not enjoy the advantage of a local monopoly the way utilities do, and because they are not thus insulated from competition, they are in a more precarious situation, hence the discount from par for their bonds. Western Refining also owns a chain of gas stations and a fleet of fuel distributors.

Operating income has in fact been on the decline for the last few years, which the company attributes to narrowing spreads between light and heavy crude, and the continued economic slowdown. Although the United States hasn’t built a new refinery in 30 years, Western Refining has in fact ceased operations at one of their unprofitable refineries, a move that the CEO claims will save them money. They have also raised the specter of asset impairments in their outlook, which is a noncash charge but makes the reported earnings drop dramatically for the quarter in which it happens.

Now, one of the things that recommended Bon-Ton stores to me was the possibility that the bondholders would do fine in a bankruptcy. I am not convinced that this is the case with Western Refining, as the convertible bonds are behind a number of secured and trade creditors in terms of priority. The firm has a total of about $2 billion in debt outstanding, and as I’ve calculated the firm has only $85 million in free cash flow to the firm. Doubling that to round out the year and capitalizing it at 10x creates a value of only $1.7 billion, although considering the difficulty of building new refineries in this country they may be entitled to a higher multiple. At any rate, it does not appear that they are as bankruptcy-robust as Bon-Ton, and considering the low interest rate coverage that is an issue that cannot be ignored.

Even so, the company has over $600 million available in borrowing capacity,  a large figure considering the firm has $2.8 billion in assets and the market cap of the equity is only $400 million. Furthermore, apart from a nonrecurring writeoff of goodwill and other impairments they have operated at a profit for the last five years. Accordingly, I have confidence that the interest on the convertible bonds is safe, at least as far as junk bonds go, and the likelihood of them being paid off at par is sufficient inducement to make these bonds attractive to any people interested in high yield debt.

The below post was a foray into the world of macroeconomics, a matter which I have no formal and little informal education in but I nonetheless feel perfectly qualified to give my opinion, just like everyone else. Now I would like to talk about issues in valuing United Online, a matter with which I have much more fluency.

I have recommended United Online (UNTD) here in the past, and I do still like their FTD online property, and I think their dialup business is capable of spinning off quite a bit of free cash flow before it ceases operations. (I am fairly indifferent to classmates.com).  By traditional valuation metrics, the stock is still underpriced, but I have noticed that their free cash flows are declining quarter by quarter, presumably as the result of the loss of some after-sale marketing programs. I think the company will eventually find a new equilibrium, but on the other hand, I wouldn’t care to place a bet on what level of free cash flow that equilibrium will be.

Last quarter, they reported net income of $14 million, depreciation of $14 million, and capital expenditures of $10 million, producing free cash flows of $18 million. If we use this figure as the basis for their full year outlook and capitalize it at a rate of 10x, we get a market cap of $720 million, which is a good 50% above the current market cap. However, the quarter before last, free cash flow was $21 million, and the quarter before then $23 million, and the quarter before that, $26 million. Fortunately, as the result of their disappointing earnings announcement and not-optimistic outlook, the company’s market cap dropped by roughly $90 million, so it seems that the market is more or less instinctively walking the price down until the company regains cash flow stability, so hopefully those of us who are waiting for that moment will not see the apparent bargain price pulled away from us prematurely.

I should also point out that for whatever reason, the company includes stock-based compensation alongside depreciation and amortization. I agree that much like the other two, stock-based compensation is a non-cash expense, and if I were one of United Online’s creditors I would support that designation. However, stock-based compensation still dilutes the equity holders, and since United Online is probably undervalued I dislike seeing so much stock being given away all the more. Stock-based compensation ran them $40m last year, $36m the year before, $19m the year before that, and nearly $15m year to date. I will say that such compensation has allowed the firm to pay down its debts (almost $30 million more last quarter alone), and although interest expense is highly manageable at approximately 1/4th free cash flow to the firm, I do like to see debt being paid down in this environment.

The bottom line in terms of recommendations is that United Online is apparently underpriced on a free cash flow yield basis, but until they can get their declining free cash flows under control, I would probably not consider investing in them without a much larger discount. In terms of book value, they’re not particularly impressive either. I believe that such an occasion, as indicated by a quarter or two of flat or increasing free cash flows, is not long in coming, and I hope that the market will not walk the price away from us bargain hunters in the meantime.