Thursday, August 25, 2011

HP Stock Gets Destroyed

Hewlett-Packard (HPQ) has seen better days, but its stock price has likely overreacted to the downside. The last time HP's shares were this low was back in 2005, when its earnings per share were about $1/year. Today, its trailing twelve-month EPS is around $4, giving the stock a P/E of about 6!

But value investors should avoid rushing in blindly, for several reasons. First, this management team is relatively unproven. The manager that grew the company's stature and earnings over the last several years was dumped last year for some improprieties unrelated to the job.

In some businesses, you don't need an all-star team running the show. For example, if you're investing in a business trading at a discount to excess real estate assets, a great manager isn't going to squeeze a lot more out of these assets than a good manager.

But in HP's field, management makes all the difference. Particularly as the company sheds its more stable PC business, HP is focusing in areas where innovation and disruptive technologies are the norm. A key success factor here is your manager.

Finally, the company's capitalization doesn't leave a whole lot of room for error. Following this $10 billion acquisition of Autonomy, the company's debt exceeds its cash by about $24 billion. This is in direct contrast to some of HP's competitors, which have billions of cash in excess of debt. If HP's execution is strong, this isn't a big deal. But if things start to go wrong in the company's fast-changing, risky space, this company won't have as much room to maneuver as will its competition.

HP might do very well in the coming years or it may not; it's difficult to say. But this isn't your father's HP, as the company is shedding its cash cows at the same time as it makes some risky bets. Only investors who can claim that this company falls within their circle of competence should consider entry. Otherwise, you may be buying something that looks cheap but is actually expensive based on its future prospects.

Disclosure: None

5 comments:

Anonymous said...

Saj,

I'd be interested in your opinion on an accounting issue. For companies such as HP, who have done a lot of acquisitions in the past, I add back after-tax amortization of purchased intangible assets to compute an "adjusted" version of reported GAAP earnings. On that basis, HP is a lot cheaper than you said. In fact, if you use their reported "non-GAAP earnings" (which also adds back restructuring charges but does *not* add back stock-based compensation), then HP currently trades an a P/E of just 4.7!

You wrote about the general subject of amortization of intangibles in a recent post, but I wanted to discuss this issue specifically for tech companies, acquisitions, and R&D. Here's my thinking. I think it's safe to assume that management does not lay off the R&D engineers from the acquired company. Therefore, the value of the acquired software or hardware IP is at least maintained over time by way of ongoing R&D. Since that R&D is expensed as incurred, it's double counting to also amortize purchased intangible assets.

- aagold

Saj Karsan said...

Hi aa,

I think it all depends on whether the company can internally generate (with the R&D dollars you talk about) new technologies that replace the acquired technologies that will go out of date. When the technologies change so fast, I think it's harder to do; so a company may have to just keep acquiring new expertise, in which case you shouldn't be adding back the amortization expense. So I guess it would come down your judgement of whether HP is adding to its expertise with acquisitions, or merely replacing expertise that is now obsolete.

Anonymous said...

Saj,

Well let's back up a step to clarify something. Let's consider a technology company that is operating on its own without doing any acquisions. Should the cost of its R&D be expensed as incurred, or should the cost be capitalized on the balance sheet? Almost all tech companies I'm aware of do the more conservative type of accounting - they expense R&D as incurred, and therefore they don't have any R&D "assets" to amortize later. But if a tech company does capitalize its R&D expense, then it will also have to amortize the previously capitalized expense over time.

But when a tech company does an acquisition, it seems to me that it's *both* amortizing the acquired R&D asset *and* expensing the salaries of the engineers it just acquired. Do you see my point? I mean, let's consider the very first year after a startup company is acquired. Obviously the company is still paying the salaries of the engineers it just acquired, and it's hard to believe those engineers are already obsolete. Well if this R&D is being expensed as incurred, isn't it double-counting to *also* amortize the acquired R&D?

- aagold

Saj Karsan said...

Hi aa,

You raise an interesting question, but I'm not sure I agree. First, I would argue that most of the difference between book value and purchase price is allocated to Goodwill, which doesn't get amortized. To the extent that there are intangibles booked other than Goodwill, I would argue that adding those amortized amounts back (for the purposes of estimating earnings power) requires a belief on the part of the investor that the company no longer needs to buy new companies to maintain its earnings power (which may be true, but must be evaluated).

The fact that the company will expense the R&D costs of the acquisition going forward is conservative, I agree, but is a separate issue IMO. Because what is being developed from now on is hard to value, that accounting treatment is likely appropriate unless the investor can prove otherwise.

Am I understanding what you are asking or am I missing the point?

Anonymous said...

Saj,

It was surprising to me when I first realized this, but it seems that many tech companies have very large quarterly amortization charges for "acquired R&D". I also thought that the difference between book value and purchase price was all considered Goodwill (which is not amortized anymore), but apparently not. For example, in the 9 months YTD for FY2011, HP has expensed $1.2B in "Amortization of purchased intangible assets". That's quite significant, seeing as their EBIT over this same period was $8.9B.

In thinking about this accounting issue, I think it's conceptually useful to separate "Maintenance R&D" from "Growth R&D". Maintenance R&D represents the spending required to maintain the current level of software/hardware sales - things such as bug fixes, new software/hardware versions, and feature additions that are required to maintain sales at the current level. Growth R&D is spending required to expand into new product areas and grow sales.

Using the above definition, I think it's clear that the expensing of R&D for acquired companies and the amortizing of acquired intangibles are *not* separate issues. If an investor believes that the ongoing R&D spent on the acquired products is at least equal to the R&D required to maintain those products (which is fairly conservative because often times growth R&D is being expensed as well), then it doesn't make sense to amortize the acquired intangibles.

Having said all that, I think I do see your point to a certain degree. In an environment of rapidly changing technology, maybe it's not always possible to "maintain" a product perpetually just by spending money on typical engineering/marketing salaries and evolving the product "naturally". Maybe sales are always destined to decline over time for the acquired products, even though the engineering and marketing people working on those products continue to get paid every year. If that's the case, then I'd agree that it's not appropriate to add back those amortization charges.

- aagold