Fortress International Group: fortitude is a virtue

by PlanMaestro

Micro caps are a nice hunting ground because they are under the radar of most investment firms. For the same reason I do not usually talk much about micro caps because I like to have the time to increase my position and do a little trading to reduce the average cost taking advantage of the low liquidity and large bid/ask spreads.

However it seems that the cat is out of the bag and the positive characteristics of Fortress International Group (FIGI) have been discovered after the recent fourth quarter 2009 conference call. The market reaction to a barely negative earnings result may be a surprise until you realize that this is a company in a cyclical industry with positive trends.

FIGI’s Management defines its industry as mission-critical IT solutions market but in layman terms what they do is construction management, consulting, and facility management for datacenters. This industry is fragmented and probably the only competitive advantage is deep relationships based on a history of giving results. Some of their competitors are Washington Group International, Dycom Industries, Mastec, Hill International, Hewlett Packard, Holder Construction, Nova Construction, Syska Hennesey, Whiting Turner and Clark Construction. The traditional competitors were engineers and contractors, but they are being displaced by these more specialized firms.

I am not very good in the timing of macro calls, and probably many investors are selective in their good memories when they think they do. However, I also think that there is no evil that lasts a hundred years. Hey, the Great Depression lasted barely a decade. So what I look for in any cyclical is the ability to grind it out. So here is what I like about FIGI:

  1. Downside Protection: the cost structure is mostly variable and they have managed to stop burning cash the last two quarters when revenues basically dried up. It has some cash, low debt and sort term maturities have been renegotiated
  2. Good Economics: There is specialized expertise and long term relationships, so even though the industry is fragmented the pool of players is stable. The pent-up demand is substantial and a bottleneck is expected
  3. Secular Growth: Datacenter needs are growing with the increasing internet traffic. Clients revenues and capital expenditures are expected to grow double digits for the foreseeable future

Over the last four years FIGI managed to transform itself from a government dependant company to a national organization with several private clients. FIGI’s customers include data center real estate developers, co-location providers, managed hosting companies as well as customers that have traditionally owned their own data center space. The following slide from Equinix (the largest colocation player) shows some examples of potential clients, not exactly FIGI’s.

Datacenters supply picked up very slowly after the 2000 internet bust given the massive overcapacity. This provided the side benefit of a more concentrated client base and more rational capital expenditures. So for the most part, there is no significant overcapacity today and it might be argued that there are bottlenecks. To give you a flavor of the favorable economics on the client’s side, this is the stock performance of REITs in this space: they include some of the greatest under reported opportunities in late 2008: Digital Realty Trust, Switch & Data Facilities, Equinix, DuPont Fabros Technology

And what are the clients looking for in companies like FIGI besides ways of reducing capex? First and foremost, expertise in ways to reduce energy costs. And this is where FIGI becomes important because technology and innovation are central.

There are three elements that provide a margin of safety for this investment

  • Variable Cost Structure: Most of its costs are consultant salaries, which can be reduced to adjust to revenues. The elements that were no variable, like NASDAQ listing costs, board size and G&A have been significantly reduced or negotiated to a variable structure
  • Low Debt: while the cost cutting measures started taking effect in Q3 2009, FIGI has some cash to navigate the issues and the $2.2 million debt maturities were successfully renegotiated.
  • Some Recurring Revenue: the facility management division has recurring revenues that while small ($5 million approximately), is growing and provided a buffer over the last year
  • Stopped Burning Cash: FIGI has been almost cash flow positive the last quarters despite depression like revenues. FIGI demonstrated its ability to survive.

However what attracted me more about FIGI is its growth potential as the need for datacenters increases. The drivers are well known: video and graphical information is getting more pervasive with access to broadband, outsourcing based on economies of scale in managing the information, proliferation of distributed applications with distributed data (commonly known as the cloud) and the outburst of data mining applications in specific industries (financial industry and database marketing). As a result, as internet traffic doubles, the demand for conditioned space to handle it increases.

In addition, there are replacement needs in this industry. Datacenters become obsolete when they do not continue to reduce operating expenses so they need to be updated. According to a previous FIGI’s conference call:

By 2010 the same survey indicated that 50% of all the data centers will have to relocate or outsource their facilities due to changing technology trends in the cost of power. The survey also said by 2010 over 70% of all date centers will implement green computing initiatives.

Another recent article in data center knowledge stated that surveys done by both APC and Digital Realty of their customers indicated over 80% of their respondents will experience a major renovation of their space in the next 24 months and this is due to technology upgrades, green initiatives and growth.

However, the credit crunch that began in September 2008 had its effect: it reduced almost to a halt the capital expenditure in these large complexes even while the positive growth trend remained intact. The reason for the recent slowdown is that the client’s business is a very capital intensive. Building a datacenter costs around $1,000/sqft, check the following slide from DuPont Fabros :Also, some of the large clients are REITs that distribute a large percentage of their earnings in dividends and are dependent on the capital markets to fund their growth capex. As an example follow the DuPont Fabros (DFT) challenges last year.

The reduced client capital expending resulted in a significant decrease on FIGI’s backlog:

And cash and cash equivalents was severely impacted too

Demand for datacenters is still growing at 13% plus per year, while supply has been constrained growing single digits in 2008 and 2009. Client’s pricing continues to go up and that is huge incentive for new capital expenditures when leasing activity starts to loosen up.

The two year lead time for the construction of one of these complexes can be critical, because the whole industry can be in a rush to catch up with pent-up demand as soon as credit is available. Also, some major companies are looking for opportunities to outsource their needs in order to reduce CAPEX. The next two quarters should be crucial to FIGI.

Now that we have mentioned Equinix, it is merging with Switch and Data. It is important to note that the consolidation of large clients is probably the most important threat to FIGI. But, as usual with cyclicals, the investment time horizon is not farther than five years and the long term effects of increasing customer’s power is not even in the same order of importance as the lack of short term demand. FIGI is not your Buffett investment, but cyclicals in general are not Buffet investments: the time horizon is short.

Next we will review the Spartan measures taken by management to reduce their cash burn and the tactical issues of where to find the idea and when to buy a cyclical.