Boring Portfolio

Boring Portfolio
Analysis of American Power Conversion's Q3 Earnings Report

By Whitney Tilson (

NEW YORK, N.Y. (November 1, 1999) -- It is with genuine regret that I wish Dale well in his new position at Legg Mason. He is a great thinker, teacher, and investor, and I learned a lot from him. I'm quite sure I speak for the entire Boring Port community when I say that you will be missed, Dale. The next manager of this portfolio will have big shoes to fill.

I'm glad that the Fools plan to preserve the Boring Port, and in the spirit of maintaining it while they choose another manager, I'd like to contribute a column analyzing American Power Conversion's (Nasdaq: APCC) Q3 earnings report. Regular readers will recall that I wrote a guest column in this space on APCC, a stock that I own personally, on September 22.

American Power's earnings report on October 21 was spectacular, and Mr. Market clearly agreed, boosting the shares 20.6% the next day. While revenues rose a modest 8.7% year over year (more on this below), gross margins increased sharply from 44.1% to 46.9% and operating expenses dropped from 24.1% to 23.1%. Net margins jumped from 14.2% to 17.5% -- a figure that would place APCC in the top 7% of the S&P 500. The result was EPS of $0.32, four cents better than consensus analysts' estimates, and 31.3% higher than last year. It was the highest EPS growth since 1997 and the 15th consecutive quarter that the company met or beat the consensus estimate. The company guided analysts' estimates up to $0.32 pr share for Q4, which would result in $1.03 per share for the year, representing 30% growth. In the conference call, APCC also said it was comfortable with estimates for $0.22 per share in Q1 2000, a 22% increase.

On the balance sheet, the news was equally good. Days of inventory fell to 104, down from 113 in Q2 and 136 in Q1. Despite APCC's continued strong growth, inventory has actually fallen 13.2% this year, as major distributors are now carrying 3-4 weeks of inventory rather than the 7 weeks they had going into this year. As a percentage of revenues, inventory was 56% in Q3, versus 74% in Q2 and 88% in Q1. The flow ratio, discussed at length in my earlier column, remained constant at 2.55, and has been in a narrow band from 2.51-2.56 for the past four quarters. While I would prefer a lower flow ratio, I'm not concerned as long as it's not increasing.

American Power has no long-term debt and its short-term debt stands at a mere $160,000, down from $15 million a year ago and $1 million last quarter. APCC's cash and cash equivalents increased to $331 million, up from $220 million year over year and $272 million sequentially. Thus, in Q3 the company earned $62 million in net income and put $59 million of cash in the bank. That's pretty impressive. For purposes of comparison, Microsoft earned $2.2 billion in net income last quarter, and cash rose by $1.7 billion. Intel, another cash generating machine, earned $1.9 billion and put $1.3 billion in the bank.

The net result of the income statement and balance sheet improvements is that trailing 12-month return on invested capital (ROIC) increased to 36.7% vs. 34.5% a year ago and 35.7% last quarter. (The company, which calculates the figure slightly differently, said its ROIC was 42%, the highest level since 1997, versus 38% a year ago.)

Lest you think all is perfect, I identified three areas of concern. The first two are minor: the number of diluted shares outstanding increased 1.5% year over year, but the recently announced 10-million-share buyback program should begin to reverse this trend. And while accounts receivable increased $46.1 million sequentially, days sales outstanding actually fell from 57 days to 56. The most important concern relates to relatively weak revenue growth of 8.7%, especially when management said during the conference call that it intended to reduce margins somewhat (back to historical levels) by passing manufacturing efficiencies and lower expenses along to customers in the form of lower prices. While this is great for American Power's customers and market share, it means that revenue and earnings growth will likely track each other more closely going forward.

Here is a chart of APCC's year-over-year revenue growth:

According to the earnings release, the slow revenue growth in Q3 was due to "inventory reductions at the Company's large distributors in the US as well as continued softness in Latin America" (where revenues declined 11% year over year). In the Americas as a whole, sales increased a mere 1%, while sales jumped 16% in Europe and 42% in Asia.

As you can imagine, the slow revenue growth was a major topic of the conference call, which you can listen to on American Power's website by clicking here (I have also posted my notes from the call on the APCC message board). Management cited two primary reasons for the slowdown. First, APCC's major distributors in North America continued to reduce their inventory levels, which meant that APCC's sales (which are recorded when APCC sells into the channel) were weak, though end user demand showed healthy 20% growth. The decline in inventories supports this argument. Second, in the summer of 1998, there were severe storms and power outages, which led customers and distributors to significantly increase their orders and inventories. This led to a spike in sales in Q3 1998 (which is clear on the chart above), making for difficult comparisons a year later.

In addition to these explanations, there are a number of other factors that point to robust future growth:

  • APCC reorganized its sales force recently, and management said that this initiative "gained traction" in the past eight weeks, resulting in 20% revenue growth during this period versus the same period a year ago.
  • Management reported excellent growth of its new high-end Symmetra and Silcon products, with 70% and 40% year-over-year unit sales increases, respectively (APCC acquired Silcon last year). The company is just beginning to roll out these products in North America, its largest market.
  • International growth remains robust and nearly limitless in its potential.
  • APCC reported no slowdown in demand due to Y2K issues.
  • APCC announced major new partnerships with IBM and Hewlett-Packard to promote and sell APCC's products. The latter partnership falls into the "if-you-can't-beat-'em-join-'em" department, as HP tried to compete with APCC in 1995 but never achieved more than 1% market share and exited the market in 1997.
Thus, I believe that the slowdown in revenue growth is a one-quarter blip, but I'll be watching this closely going forward.

To read Mr. Tilson's three previous guest columns in the Boring Port and other writings, click here.

The Fool is hiring. Answer the call.

 Recent Boring Portfolio Headlines
  10/30/00  American Power Conversion's Ugly Earnings
  10/23/00  Cisco's Formidable Challenge
  10/16/00  Cisco, Apple, and Probabilities
  10/09/00  Perils and Prospects in Tech
  10/02/00  Learn From Mistakes
Boring Portfolio Archives »