Comfort Systems USA Inc (FIX) 2004 Q4 法說會逐字稿

Welcome to the Comfort Systems earning release conference call. At this time all participants are in a listen-only mode. After the presentation, we will conduct a question and answer session. [OPERATOR INSTRUCTIONS] Today’s conference is being recorded. If you have any objections, you may disconnect at this time. I would now like to turn the meeting over to Mr. Gordie Beittenmiller, Chief Financial Officer. You may begin.

Thanks, Dori. Good morning, everyone, and welcome to Comfort Systems USA’s fourth quarter earnings call. At the outset, we want to remind everyone that our comments this morning, as well as what we issued in our press release, contain forward-looking statements within the meaning of the Private Securities Litigation Act of 1995. What we say is based on the current plans and expectations of Comfort Systems USA. Those plans and expectations involve risks and uncertainties that could cause actual future activities and results of our operations to be materially different from those set fourth in our comments. A more extended list of specific risks is detailed in our 10-K and our press release covering these earnings. On the call this morning are Bill Murdy, Comfort’s Chairman and CEO, and Tom Tanner, our Chief Operating Officer. Bill is going to open our remarks.

Thank you, Gordie, and thank all of you for being on the call this morning. We have lots of people who have been with us before on these calls and some new ones.

We’re pleased to report this morning that we capped a very strong 2004 with a very solid performance in the fourth quarter. Our GAAP earnings per share came in at 5 cents a share, compared to a loss on a GAAP basis of 12 cents a share in last year’s fourth quarter. For the full year, GAAP earnings per share came in at 27 cents versus a loss of 17 cents last year.

Excluding items that are not typically seen in every period, such as goodwill impairments in both years and restructuring and credit agreement termination charges in ’03 and further applying a higher more normalized tax rate of 42%, to results in both years, net income in the fourth quarter more than tripled from 1.1 million in ’03 to 3.7 in the fourth quarter ’04. As did EPS going from 3 cents in last year’s fourth quarter to 9 cents per share in this year’s fourth quarter.

Calculated on a same basis, full year net income increased about 150% to 12.6 million in ’04, compared to 5 million in ’03. Full year EPS determined on this basis was 32 cents a share versus 13 in ’03.

We, in the fourth quarter, posted a respectable 7% increase in revenues and significantly added to our backlog with increases of 11% from the third quarter this year, and, indeed, 42% from last year’s year end. We started ’05, thus, with a backlog at a record $573 million. Gordie will have some more comments on backlog later, and to include the fact that not all of our work passes through what we measure as backlog.

We are seeing recovering activity in most of our markets, and our operations are capitalizing on those opportunities. Cash flow arena—we produced excellent free cash flow in the fourth quarter, bringing our full year total to about 22 million. At year-end, our cash net of our debt position on our balance sheet increased to 24 million as of year-end.

We are also pleased to have reached an important milestone in today’s heightened accountability environment for public companies, consistent with expectations and requirements of the Sarbanes-Oxley Act ’02. We conducted an extensive evaluation of the Company’s internal controls, which led to our conclusion, and our auditors concurrent, that these controls were operating effectively. We, by the way, estimated that the cost of this was about 2 cents a share, or about 6% of earnings. What we can say is that stakeholders can have as much confidence as ever in the quality of the information that we report to you about Comfort Systems.

Moving on, we did make a comment in our release—former release—about the first quarter of ’05. Frankly, the unprecedented rain in California that we all have heard about has had a substantial effect between our units located in that state and the presence we have there from units based elsewhere. We have a notable amount of work in California. These operations have experienced both job interruptions, as well as there have been reduced levels of new work coming to the market in both of these developments, stemming largely from the weather impact. We think this is temporary, and our backlogs in California remain at historically high levels.

We have also experienced some challenges associated with the very active multi-family operations area. It’s a very active market for us, both garden and, especially, high-rise multi-family activities. That’s been very strong over the last year, and, as you’re aware, we have some very strong well-established expertise in this area. As a result, multi-family work actually comprises about 20% of our backlog.

Our largest operation, based in Houston but operating all over the country, including California, focuses on this sector. And we have recently seen a number of customer and/or weather driven schedule delays in that book of business.

We can often respond to a lightening of schedule by reducing our costs. But this particular operation has a record backlog, which is currently scheduled to ramp up in the spring, and we have very strong prospects, as well, continuing into 2006. We, thus, need all of our current resources to meet these opportunities. Consequently, we are carrying some current costs and resource levels through the first quarter, and recognizing lower levels of current utilization of efficiency. But we’re doing that to ensure that we can execute on a very good book of business going forward through ’05 and, indeed, into ’06.

We feel the effect of these factors in the first quarter, traditionally our lowest activity, will likely result in our first quarter results being off from last year’s relatively strong results. That said, we would note we continue to see very healthy indications in our markets and throughout our operations, the most recent non-residential activity, statistics, as reported two days ago, show a continued strengthening in activity levels in non-residential sector, with the last two months reported December and January, showing high year-over-year gains over a couple years. And the major HVAC manufacturers have all reported improving trends in their commercial market and outlook. We have a high level of backlog, which, of course, supports these statistical objective numbers, and we have positive margin developments in that backlog, which Gordie will discuss in a minute.

And our productivity and execution efforts, which we continue to work on, are continuing at full pitch. Tom will discuss that shortly.

As a result, even with lower first—fourth—first quarter, we still expect to produce higher revenues, improve margins and year-over-year growth in results for 2005 as a whole.

One last note, as previously announced, in January we brought our first new company in to Comfort in over five years, with the addition of an excellent and well positioned team at Granite State in New England. We’re very happy about that operation joining Comfort Systems.

In summary, overall, we have a very positive outlook on where we are right now and what we can do over the next few years. Comfort Systems, we believe, has never been stronger or better positioned for continuing long-term performance.

With that said, I’d like to turn the mic over to Gordie, here, for comments on the financial side.

Thanks, Bill. Bill noted our revenue growth for the full year as we kept pace with re-emerging industry growth.

Looking into our book of business a little further, revenue remains well diversified. Multi-family was 15% of our mix, office buildings 14, healthcare 14, manufacturing 12, retail 9, government 8, education 8, the rest of our work spread among smaller miscellaneous sectors.

Our backlog indicates we will see increasing activity in the institutional markets of government and education and healthcare, as well as multi-family.

Looking into our book of business from a different angle, our average project size, as of December 31, was $240 thousand, reflecting over 4,700 projects in progress. We had two projects at $20 million, six projects between $10 and $15 million, fourteen projects between $5 and $10 million and 154 projects between $1 and $5 million and, finally, more than 4,500 projects under a million.

Almost two-thirds of our backlog value is in projects under a million dollars. This distribution of activity demonstrated a stronger diversification in our book of business than many observers appreciate and improves our ability to work through the ups and downs, currently generally up. So the broad non-residential building sector.

Our gross profit percentage increased in the fourth quarter on the strength of lower safety and risk management costs and notable execution improvements at a number of our operations. Net of lower margins at higher volumes in our multi-family operations in the beginning of under absorption of resources that are larger multi-family units. As Bill mentioned, it’s schedule slowed ahead of a larger activity ramp scheduled for later this spring.

In addition, we decreased our project and service overhead in gross profit by 1.3% year-over-year. Looking to gross margin indications in our backlog, we see some positive signs. Our aggregate gross margin and backlog was down slightly, 20 basis points sequentially. And it was down 100 basis points year-over-year. However, if we stratify margins somewhat, a different picture comes into view. Through the second half of 2004, and in the fourth quarter, we saw increases in the amount of our backlog comprised by projects at the $5 million size and greater. This was offset by a decline in backlog for projects between $1 and $5 million. And we saw a significant increase in backlog for projects less than a million.

The increase in the larger projects adds longer scheduled work into the backlog at generally lower margins, more conservatively estimated margins. This has had the effect of lowering our aggregate, but somewhat more than is usual, because we are booking some of this work further in advance than we typically do as market activity rebounds. In other words, we have essentially laid more of a 9, 12 and 15-month foundation of project work into the backlog somewhat earlier than usual. And, again, this project work comes into our backlog at lower margins.

Looking at this another way, our 6-month backlog—in other words, work scheduled to be performed over the next two quarters—was essentially unchanged from September 30 and up only 8% from last year-end. What that indicates is that the substantial majority of our backlog increase came in the nine-month and greater backlog. Again, at this early date, that comes in somewhat larger pieces and has somewhat lower margins, more conservatively estimated.

Looking more closely at our six-month margin, which reflects a more typical mix of project sizes and work at all stages of completion, we see margins that are up 20 basis points from last year and up 50 basis points sequentially. We would expect when we approach the third quarter of next year, the foundation of larger project work we have now put in place will be overlaid with the typical shorter and faster turning smaller project work, which generally comes in at higher margins, and as current activity trends continue, may begin to show better pricing.

Now, as we always note, approximately 40% of our revenues, including service work and quick-turn projects—very fast projects—is not captured in any particular quarter-end backlog cut off. So backlog indications, while general directional indicators, do not guarantee future results. And price compensation is still a factor in our markets. However, we are encouraged by the margin trends in our near-term backlog and believe they suggest the beginnings both of some pricing progress as well as improvements in projects execution and productivity.

SG&A was up 6% year-over-year and 7% sequentially, primarily as a result of increased accounting fees, mostly associated with Sarbanes-Oxley compliance, higher incentive compensation and increased bad debt reserves associated with certain contested receivables. While we have indicated that we would begin to see SG&A increases again as our growth returned to our top and bottom lines. It remains our goal to hold increase in our SG&A project and service overhead to less than the growth in our gross profit.

Netting out gross profit in SG&A, Bill earlier noted our strong quarterly gains in income and earnings per share. Our operating margins, and ongoing operations, virtually doubled from 1.6% to 3.1% for the fourth quarter and gained 130 basis points to 2.9% for the year as a whole.

We did record a non-cash goodwill impairment charge of 3.3 million in the fourth quarter, comparable to a similar charge of 2.7 million in last year’s fourth quarter. This charge related to our conclusion the profit levels that three of our smaller operations were likely to remain lower for an extended period of time, as compared to the levels these units earned when we acquired them in the late 90s, and stronger in overall market conditions.

We would note that the new accounting rules relating to goodwill that went into effect in 2002, contemplate periodic impairments of goodwill for business units that have declined in value, while allowing no recognition of increases in business unit values that may have occurred. As a result, we may record additional goodwill impairments in future years, even when the aggregate value of our business units, and our Company as a whole, may be increasing.

On the balance sheet, we posted a significant increase in free cash flow for the quarter, bringing our full year free cash flow to 22 million. Our cash net of debt position widened to 24 million, and we remained near the upper end of our historical range of net over billings. Our balance sheet continues to be a growing asset for us, providing us both strength and flexibility.

Finally, on the surety front, we’ve added an additional surety and significantly increased our indicated surety capacity to support our ongoing growth. We have maintained close and constructive relationships in the surety markets and, as a result, we have never had a bond request declined, nor have we ever opted not to request a bond based on indications that it might be turned down. We believe the strength of our bonding relationships has been a worthwhile competitive advantage for us.

That’s it on financial, and now I’d like to introduce Tom Tanner, our Chief Operating Officer.

Thanks, Gordie. Good morning, everyone.

On the operations side, 2005 is clearly all about improved execution. This includes the profitable execution of our record backlog, this includes improved execution within all of our service divisions to increase both revenue and operating income, and this includes improving the execution of all aspects of the operations in our under-performing companies.

Relative to project execution, we have completed our initial productivity improvement training as kind of our operations in 2004. We will have ongoing follow-ups of the training during 2005. During Q4 we began to see positive results from our efforts in terms of more consistent and improving project profitability. We are encouraged we will see a continuing improvement this year as productivity improvement techniques become more engrained in each company’s culture.

We are currently conducting initial productivity training, some of which we started in Q4, at 12 more of our operations with five more companies scheduled to begin training later in the year. The implementation of the productivity training has been accelerated because of our large backlog. We continue to expect these efforts will be a key factor in maintaining and improving our gross margins during ’05 and ’06.

Another example of our commitment to improving our project execution is that we are completing development of a weeklong project managers academy program that will provide industry specific training for our project managers. We expect this additional training will be provided to approximately 100 individuals during the second quarter.

Relative to our service divisions, in an effort to accelerate the improvement within our service divisions, we have added two key individuals—Jim [Shaffen] and Fred [Worse]. Both of these individuals came from the preeminent HVAC service franchising company in our industry. They have extensive experience in our business, and their roles at Comfort will be very similar to their roles at their former employer. They will team with Virgil [Harris], who joined the Company in the third quarter of last year, to develop and implement a more standardized service offering, to increase service revenue and to improve the overall profitability of all of our service divisions. With this increase focused on our service divisions, and our continuing traction in our national and regional accounts activities, we are encouraged about our prospects for profitable growth and the very important service segment of our business.

As we noted in earlier quarters, we continue to experience operating losses in some of our operations in 2004. Importantly, both the number of units with operating losses and the aggregate amount of the operating losses, were substantially reduced in 2004 as compared to 2003. However, we must really improve our overall execution of the companies in 2005 to achieve significant improvements in their results.

We continue to carefully evaluate the management in these companies, and we’ve made management changes early this year to prepare for a much-improved 2005. We believe that most of the problems that plagued these companies in ’03 and ’04 are now behind us.

With a strong and well-positioned backlog, we now have a number of operations that are sold out several quarters in advance. In these, and other very active operations, we are steadily increasing our selectivity and pricing in accepting new business, while working on adding additional labor and project management talent to support further growth. Our goal is to successfully manage our project schedules, while avoiding biting off more than we can chew.

2004 was a very rewarding year for Comfort Systems, as a whole. Having navigated through the recession of ’02 and ’03, we will prepare to take advantage of the improving 2004 economy. With our record construction backlog and our growing service business, we are very excited about what the balance of ’05 and ’06 hold for us.

With that, I’ll turn it back to Bill for wrap up and Q&A.

Good. We definitely want to elicit some questions. Thanks, Tom. I’d like to conclude with one statement in recognizing the outstanding contribution of Comfort Systems people across all of our operations. 2004, of course, was a year of superb improvement for the Company, operating more safely and effectively than any year in our history. And that result was delivered by the more than 5,800 people that make up our team. We thank them all and look forward to building on 2004 with an even stronger 2005.

We would like to open it for questions. There’s a fairly large group out there, but feel free, please, to indicate that you have a question, and we’ll get right at them.

Thank you. At this time we will begin the question and answer session. [OPERATOR INSTRUCTIONS] Our first question comes from David Yuschak with Sanders Morris Harris.

Thank you. Congratulations on your fine quarter. Just a couple questions on the backlog. Just pretty impressive quarter-to-quarter growth, and generally speaking, things should be winding down and the bookings really begin to pick up early in the year. But that kind of popped from the third quarter to the fourth quarter. Given the kind of level of revenues you’ve had, could you just give us a little more clarity as to where that’s happening? Kind of what projects that may be happening in? What maybe you guys are doing in the market place to be able to show that kind of impressive growth, because I don’t know that anybody else in this space had that kind of experience in backlog growth in your space.

Let me comment, briefly, on a couple of the sectors, Dave. And I’ll comment on what might be the general underlying trend. It would appear to us that particularly when you look at the non-residential information that’s come out, forecast and the expectations, that their has been a growing degree of confidence in the non-residential sector, which is a lagging sector, that the improvement in economic conditions might be here to stay for a little bit. And people are getting more comfortable in making building commitment decisions, both in new construction and significant replacement. And that cycle, then, tends to get a little bit of momentum as people start booking projects and using capacity. And users and [TCs] begin thinking, Gosh, I’ve got to schedule things a little further out. And I think we’re seeing a little bit of that in terms of growing backlogs.

In sectors we continue to see a lot of activity in the multi-family sector. That’s more than 20% of our backlog. There’s a lot on the board, still. It looks like for our particular operations with their reputation and expertise, as that strength will continue throughout ’05 and into ’06, as far as we can see at this point. The fact that money is still relatively inexpensive, certainly helps that.

In the other sectors where we have seen continued steadiness and strength for us is in the institutional sectors—government, healthcare and education. Office buildings and manufacturing tend to be up and down. Office buildings were up for a couple quarters for us but have backed off a little bit. But institutional multi-family seems to have ahead of steam for us in our markets right now.

Now are you guys doing anything different, though, that you’d be able to capture that kind of backlog growth that others maybe aren’t?

Once thing that has certainly helped us, coming out of the downturn, has been our steady surety situation. We have been able to bond everything that has required a bond. And in the multi-family sector, in particular, when you look at some of the project work there, some of the high-rise work, our bonding capacity has been a really competitive advantage there. So that has been a factor for us. And you do use a little bit more bonding. It is more of an expected practice in the institutional sector, where more of our strength has been, as well as multi-family. So that’s certainly helping.

On pricing, you’d indicated that with resources getting tighter, is it possible that we can begin to see pricing improving throughout 2005, given this kind of strength, but as you see it in the non-residential area?

David, this is Bill. We think so. We have not seen a lot of evidence of price increase traction yet, as that may indicate that the general sector capacity is still not fully utilized. We’re getting a little more than our fair share, as was reflected in your question and Gordie’s answer. And we’re going to run up against resource constraints, and that’s why we’re focusing on taking the best work. But we just haven’t yet seen the pricing traction. But intuitively, and factually, it’s a supply-demand situation at the end of the day, and I think at some point we will see that traction. Tom, can you comment on that?

Well, typically, private companies, at the beginning of the year, have the sense that they need to build backlog, and oftentimes go out and take work in the first quarter to try to almost sell themselves out for the rest of the year. It’s a psychological thing. And we wait until later in the year, and as these projects come on board, I believe we will be able to get some price increases, because we will still have capacity and resources available that a lot of our private competition will not have.

And then just one last question, and I’ll back to the queue. On the cost side of it, Gordie, you’d indicated that when we take a look at the numbers, SG&A was about 2 million versus the third quarter, basically on the same kind of revenue that you produced in the third quarter. Could you give us maybe a better breakdown on where that might have been? Because you did indicate the risk in some receivables you took a charge on and bonuses and a few other things. Could you maybe give us a little clarity on that and what that number might be going ahead? What kind of growth rate you may be able to see in SG&A factoring in all the things that you need to for 2005?

And then for you, Bill, you’d indicated you’ve been building up some extra resources. Is that primarily showing up in the first quarter that may be limiting somewhat what you’re also doing in the way of potential earnings in the first quarter as well?

Let me comment on SG&A first. You noted two of the three things we’d mentioned in our remarks on SG&A. One of them is incentive compensation. We had certainly a better year in ’04 than ’03, so incentive compensation has increased and that, obviously, gets firmed up at year-end and in your fourth quarter. Although you accrue it throughout the year. Going forward, as we mentioned, we expect to have a better ’05 than ’04. That may or may not result in increased incentive compensation, depending on where those improvements are and how they distribute out. It’s reasonable to expect that that level of incentive compensation will continue, but it shouldn’t show the same increases, necessarily, in ’05 that we had in a bigger turn around year of ’04.

Sarbanes-Oxley and accounting fees were a bit factor in SG&A. We like to think that they will come down a little bit going forward, now that we’ve gotten ourselves and our auditors through the first year of learning curve and trying to safe-side everything that we do to ensure that we can withstand any scrutiny as to how we did our Sarbanes certification. But too soon to tell whether Sarbanes costs will come down. We think they will.

We did record some bad debt reserves in the fourth quarter on two particular situations that amounted to approximately $0.5 million in the aggregate. We don’t expect to have that kind of special additional bad debt reserves in every quarter. So that is something that should not be recurring at that same level.

And then, I guess, you had a question regarding resources and the effect on first quarter.

Yes, I think there certainly has been some, Dave. We have a huge backlog that is going to ramp up with better weather and some of these project delays ending. And we just can’t turn this on and off. Even variable costs—certainly semi-variable, really—and so we probably are carrying some SG&A that we wouldn’t otherwise have here.

That’s probably showing up in the first quarter, is kind of what you’re saying, because the fourth quarter didn’t have much of that in it.

That I think is right. We haven’t really done a whole lot of analysis to differentiate. But I think you’re right.

Okay. Thanks.

Our next question comes from Craig Irwin with First Albany.

Hi, guys. My first question is regarding acquisitions. We really like the Key Span acquisition in the quarter. It’s nice to see such a good value accretive acquisition move in. Can you talk a little bit about the potential for similar acquisitions over the next year, and what your commitment is regarding acquisitions at the moment?

Thanks for your question. It’s a logical question, our having returned to acquire something after five years. But I’d like to emphasize at the outset that we really don’t have a program. We are not going to become an acquisitions platform, necessarily. We do recognize the acquisitions of the right operations with the right functions and the right fit characteristics as part of our future growth. Candidly, our first test in an acquisition is whether that operation really, truly wants to be a part of Comfort Systems. We are not out trying to influence persons beyond the normal to bring their operations into Comfort Systems. We want a meeting of the minds on that issue to start with. Can we buy things reasonably—don’t want to say buy—bringing to Comfort Systems reasonably. The market has much changed since that of the late 90s. Expectations are not as high, capital has learned a bit here—we’ve learned a bit—we’ve learned a lot, actually, about acquiring and assimilating—so I don’t think we will see anything—I wouldn’t want to set up the Granite State acquisition as the absolute benchmark for multiples of anything. We will seek to do that, obviously, but I think I’m back to my first comment. It’s not a program. It’s part of our growth strategy, and we want to add things which fit functionally and fit with our organization. But not necessarily to achieve perfect geographic coverage. That doesn’t get you much, really. We have national customers, for instance, who have total geographic coverage, and we can service them where we are not, physically, with affiliates of ours. So we will acquire, yes, we will acquire prudently and carefully.

Okay. That’s very fair.

My next question was about the 1Q guidance. I wanted to dive down into this a little bit deeper than what you already mentioned. One of the things that jumped out at me is only two of your roughly thirty units with revenues over $10 million from California, and you did mention this Houston unit, but can you give us a little more color here and help us understand how this book of business in California can contribute to results that are down year-over-year?

Well, California is not the only factor in the first quarter expectation. It is one of two significant factors. We have operations based in California, and, as Bill mentioned, we have our Houston, and others, of our companies have operations in California. And the weather impact in southern California is a kind of impact that they’re not accustomed to out there. If you had a lot of snow in the northeast, people are somewhat accustomed to that. If you have a hurricane, people are somewhat accustomed to that in places that it is. So this has had a somewhat larger effect on the market, people stepping back and trying to assess what’s going on, delaying projects and not bringing new work to the market. But that is only one factor in the first quarter.

The other factor is at our largest market—largest multi-family operation—the gap in the schedule a little bit. One might ask, Well, why didn’t you schedule better? Well, when we put things onto the books, we do the best job we possibly can to lay in the schedule rationally. But a lot of times changes in individual situations and individual jobs, customer changes, or decisions will move things around in the schedule, and that’s exactly what happened with our largest operation here coming into the first quarter. They have a very strong ramp-up scheduled to begin in the second half of March. But right now it is off a little bit.

Both of those factors contribute to variance from expectations for those units. And when you put that up against the first quarter, which is traditionally our lower seasonal quarter, with lower levels of absolute activity and absolute revenues and profits, has a little more of an impact than if it had occurred in another quarter. Based on that, that’s why we’re suggesting that we will have a decrease in our results, comparatively, in the first quarter. But still believe, given the strength of the backlog and everything else we are seeing in our operations and markets, that ’05 will be an up year, of course, again.

Okay, great. And then if we could just touch on the backlog. One of the numbers that I missed in your prepared comments was the six-month backlog you gave the sequential and year-over-year comparisons. Do you have those numbers again?

We didn’t actually state what precisely the six-month backlog number is. We did say that it was essentially unchanged from what it was at September 30, and it was up 8% from last year.

Okay. So the six-month backlog is growing, more or less, in line with the market?

Right.

Okay. Excellent. And can you comment if there’s been much of a shift in the book of work between design build, plan of spec and whether or not you think a shift is likely to occur over the next year, based on your backlog?

Let me give you some specifics, and then Tom can talk about the direction more meaningfully. Design build is up to almost 40% of our project work and had gotten as low as 30% in the tough economy—or in the low 30s—and it’s now back to almost 40% of our project for the business.

We find in the multi-family arena a great deal of design build and/or design assist of activity. And in many of our markets, our companies are able to selectively go after opportunities where limited competition, may be two or three bidders, but, again, it’s design assist in value engineering on these projects. So we’ve seen a positive turn away from our plan of spec, or towards this design building. We see that improving as we move through ’05. And even some of the clear—what looks like plan of spec work is really design assist work, where our companies have gone in and, either prior to or after the fact, helped the owner get the project within budget and has allowed us to secure some of this work.

Okay, great. And then, Gordie, my last question is about the goodwill charge. I was hoping you could give us a little more color there? If you have a number of units that were affected or specific geographic region, or if you had any particular comments you could share?

These are three units, relatively smaller units, and really that’s the common denominator of them. And there’s not much commonality beyond that. You look at an operation, and you look at its current levels of profitability and cash flows, and you look at its market position and where it can be expected to go in the foreseeable future. Built on valuation models and that, compare it against the goodwill that you’ve got on the books, and each of these three units is profitable, but they’re just profitable at lower levels that look like they will stay at lower levels, as compared to where they were in the much stronger markets that they had in the late 90s.

Okay, excellent. Thank you very much.

Our next question comes from Rich [Wasalowski] with Sidoti & Company.

Good morning. I just have one question for either Bill or Gordie. Can you explain why the larger jobs that you guys look generally, which presumably have greater risk, are in lower margins than the smaller short-term jobs?

The answer to that, Rich, is the larger projects typically have large amounts of equipment, subcontract work, which is inherently less risky than the labor portion of our business. So we typically mark up labor to a greater extent than we do those other components of a job. And that’s why the larger projects have lower margins, because the less risky components are a larger part of those projects.

When you look at the way people make decisions, as well, a larger project to a contractor represents more use, or coverage, of his overhead—more use of his capacity. In one fell swoop, you can book business that might have taken five other fell swoops to book in smaller projects. So while there is certainly more execution risk, there is less risk when you get a larger project of knowing where your business is going to come from—of having work to do. And as a result, that also attracts, generally, more competitors—or more people interested in going after such work. And those factors play into why margins tend to get a little narrower on those jobs.

Thank you very much.

Our next question comes from Arnie Ursaner with CJS Securities.

Hi, this is Bob [Labick] for Arnie Ursaner. One quick question—you’ve given us a lot of great details on the backlog and the margins, and such. Can you help us relate this to your stated goal of 4% EBIT margins? And should those be achievable in 2005?

It’s certainly our stated goal. It’s going to be candidly a little difficult to get all the way there in ’05. We have, certainly, the potential to get very close. But we can’t guarantee that for sure, obviously. It’s a goal—we think it’s not only a goal that we can achieve, but we can sustain as we continue to work on the Company. It has to do with productivity, efficiency, eliminating loss operations, etc. Now we’re talking aggregates here. We have plenty of operations that achieve that kind of operating income level, and more. Much more. But remember we’re talking aggregates here, so we’ve got to eliminate the weak activities, and we’ve spent a good amount of time doing that. I think we’ve got at least in shape to help us with that, as opposed to hurt.

Okay, thanks. Good luck with that.

We do have a follow-up question from David Yuschak with Sanders Morris Harris.

With the construction activity picking up like you had indicated, and everything points to non-residential construction quality year for activity in 2005. Could you guys give us an update on the national accounts effort? Is that going to get crowded out because of your activities of the new construction—given the prospects of how that performed last year versus 2003? And then just give us an idea what your focus may be there for 2005, given everything else in record backlog and everything else that you may need to keep this thing on track as well.

Dave, first to answer to one part of your question. It certainly is not going to get crowded out. We’re not going to allow it to get crowded out. We’re not going to allow either national accounts work or, more generally, service work, maintenance repair, etc., get crowded out at all. In fact, we’re going to be pushing on that. Tom mentioned that the addition of some very experienced and talented people in the service area to continue to work on that. That more relates both to the service activities at our individual operations and the service activities that we perform for national accounts. We have landed some new accounts. We’ve added a couple of thousand sites recently, and are up to over 10,000 sites. It’s a painful process in the sense of long sales cycle, because you basically working with a company to change the way it does its internal maintenance, etc. You’re usually there having to let people go, start having to change vendors if they’re going outside, and aggregating to one vendor. So it’s a lot of process change there that makes it a long sales cycle. But we’re certainly not going to let it get crowded out. Tom?

I agree totally. We’ve had a significant increase in revenue in ’04 from ’03, and our run rate, starting in January 1 of ’05 is a significant improvement over the actual revenue in ’04. So we are pleased with the results in that business, and we will continue to focus on going forward, as Bill said.

Would you mind sharing with us what that kind of rate may be or what your actual in dollars was in 2004? If you wouldn’t mind? Or if you don’t, that’s fine, too.

We, in that operation, had revenues between 20 and 30 million in actual for ’04. And our run rate, as we start the year, as Tom mentioned, is up noticeably from that, but that’s probably not a number we should disclose at this point.

Okay. Thanks.

[OPERATOR INSTRUCTIONS]

I think we don’t have any further questions in the queue here. Thank you all very much for your participation in the call. We are looking forward to our ’05 here, and we thank all of you for your attention and support of Comfort Systems. Thank you.

Thank you for joining today’s conference. That does conclude today’s call. You may disconnect at this time.