Description

Bennett Environmental (BEL) is a wonderful and simple business with Microsoft-like operating margins and incremental returns on capital approaching 100%. BEL’s industry is a de facto legislated oligopoly. BEL has been unduly punished for a temporary hiccup in their business. Most importantly, BEL will likely more than double its earnings power in the next several months, and BEL shares can be had for six times its these potential earnings.
BEL owns a 100,000 metric ton per year incinerator used to treat contaminated soils. Soils too badly contaminated with hazardous materials to dispose of in landfills are incinerated to destroy contaminants before being disposed of in landfills. Incineration is the most effective method of treating contaminated materials. BEL is a Canadian company that trades on the Amex under the ticker BEL and on Toronto under the ticker BEV. All financial numbers in this write-up are in Canadian dollars, but share prices are U.S. dollars unless otherwise specified.

Plant Shutdown
First a word on the snafu that presented this opportunity. BEL’s revenue base is understandably lumpy for two reasons: they currently only operate one plant, and the projects BEL works on are by nature enormous. Two projects dominate BEL’s revenue stream at this point. The largest, Phase III of the Federal Creosote project (a Superfund site administered by the U.S. EPA), was around 75% of 2003 revenue by our estimates and was 74% of 2002 revenue. This project should be completed by 12/31/05. BEL is currently bidding on Phase IV of Creosote, which is basically in the bag according to management. There are 200 Creosote sites left in North America, which will provide work through 2008. The second, the Saglek project administered by Canada’s Department of National Defense, was around 20% of 2003 revenue by our estimates and was 11% of 2002 revenue. The Saglek project will be complete in 2004. There are 30 sites left under the program in which Saglek is included. These 30 sites will be completed at the rate of three per year, leaving 10 years of work under this program. It should be noted that BEL has to bid for each of these projects, and none of this business is guaranteed.
The controversy lies in the plant’s Q2 shutdown. In March BEL ran out of soil inventory and two large customers, GE Canada and Creosote, delayed shipments leaving BEL without a supply of soil to process. As a note, BEL keeps some soil on hand to mitigate the volatility of deliveries, and since a large portion of their soil comes from Canada, to ride out the winter when excavation is impossible. GE Canada delayed shipments beyond the expected delivery date due to weather. Creosote delayed shipments due to excavation difficulties – specifically the local power company was too backlogged to move powerlines on the site. All other Canadian projects were delayed for the same reason as GE. All of BEL’s other projects were scheduled for May delivery at the earliest. BEL expects to restart the plant in mid-May, when both Creosote and GE will begin shipping again.
At any rate, the long-term investor will largely disregard these troubles as short-term. The same situation occurred in Q2 2002 when two large customers delayed shipments. BEL’s stock fell from $12 to $5 and subsequently recovered to $10 in about a year. Going forward these situations will be inevitable, but three things will mitigate this risk. First, BEL expanded its storage facility at its current plant in St. Ambroise in mid-2003 from 10,000 tons to 40,000 tons. Second, as industry capacity utilization increases, so will backlog (more on this later). Lastly, adding a second plant will diversify their revenue stream (more on this later, too).
As an aside, especially since I’ve libeled a couple of ideas on VIC for the issue of customer concentration in the past, I’ll just say that I don’t believe customer concentration is a deal breaker in BEL’s case, although it’s obviously an issue. First, BEL’s customers are signed to contracts specifying the aggregate deliveries due, though obviously timing is a variable. Furthermore, we theorize that the demand side of BEL’s industry is not as anti-competitive as the supply side. BEL currently has a $1 billion sales funnel and a 70% success rate in bidding. The completion of any one project will likely be replaced without much trouble if further work on the site isn’t gained in a renewal contract. Lastly, BEL currently has a backlog of 400,000 tons (over four years worth of production for one plant) under contract. Based on our own rough estimates we believe roughly 75% of that backlog is associated with Creosote.

Competitive Conditions
We believe that BEL’s economic windfalls are primarily attributable to over-regulation. The industry hasn’t added new capacity in about 10 years. This is understandable when one considers the political wrangling necessary to open a new plant. If one wanted to build a new incinerator today you could expect a five-year wait for permit approval. Conditions are even worse in the U.S. Regulations passed in 1998 limit the disposal of contaminated materials in landfills, which essentially created the incinerator industry. Industry capacity utilization is around 85% today. At the same time, tighter emissions standards are making construction of new incinerators tougher. Reference BEL’s difficulties in getting its two new plants online for a real-time account of this fiasco. Note that BEL’s Belledune facility has been placed on a relative fast track (permitting having only begun in mid-2003). This is attributable to another competitive advantage. BEL has a near-perfect track record on emissions. BEL’s competition has a history of violations. This expedites the permitting process for BEL.
Other regulations put BEL at an advantage as well. Namely, the U.S. imposes restrictions on the number of deliveries a plant can accept per day. BEL is not subject to these restrictions. This means that excavation doesn’t take as long for BEL’s customers, thereby reducing costs.
Legislative barriers are only one of BEL’s competitive advantages. We believe the next most important advantage is BEL’s location. First, as mentioned before, Canada is a much friendlier place to run an incinerator than the U.S. Environmental laws are typically more lax and they lag the U.S. in developing new policies. Quebec only recently passed stricter landfill laws similar to the ones passed in the U.S. in 1998. A good proxy for the difference in the regulatory climates are the MACT standards recently passed by the EPA governing incinerator emissions. Clean Harbors, BEL’s largest competitor, recently spent $20 million to bring its 120,000 ton Deer Park, TX facility to compliance. This is as much as BEL will spend to build its new 130,000 ton facility in Belledune, New Brunswick. As a side note, the MACT standards are expected to further worsen the supply imbalance by putting 100,000 tons of captive incinerator capacity out of business. Captive incinerators are owned by companies for their private use. The closure of this capacity would drive more business to public incinerators such as BEL’s. BEL is subject to the MACT standards since they have U.S. customers, but both constructed plants are MACT-compliant through 2007 with no improvements necessary.
Another big competitive advantage for BEL relating to its location is transportation costs. Incineration is somewhat of a regional business. BEL management estimates that roughly half of the contaminated sites in North America are in the northeast region of the U.S. Transportation costs from New Jersey to Quebec are roughly $75 per ton. From New Jersey to Texas would cost roughly $140 per ton. This means that BEL can charge the same price for a project, but less of it gets passed along as zero-margin transportation revenue. Hence, BEL is a low-cost provider.

Growth
The St. Ambroise plant processed 76,000 tons in 2003 and 55,000 tons in 2002. Simply ramping up to 90% capacity (90,000 tons) would add about $4.9 million to after-tax cash flow per year:

Rev./Ton $650
Incr. Tons 14,000
Cost/Ton $120 (incremental)
After-Tax CF $4.9 mill (taxed at 34%)

Our model assumes 90% capacity going forward. Of note, before the Q2 situation management was guiding for 92% utilization in 2004. Further, BEL achieved approximately 86% utilization in Q1 if we add back 1,500 tons (12.5 tons/hr x 24 x 5) for the five days the plant was shut down during Q1. Also note that 76,000 tons equated to 87% utilization in 2003 since BEL was only permitted for 87,600 tons (10 tons per hour) in that year. 55,000 tons equated to 79% utilization in 2002 as they were permitted for 70,000 tons (8 tons per hour) in that year. Recall the incident in Q2 2002 similar to today’s situation, which produced such low utilization in 2002. For a more thorough explanation of permitted versus physical capacity, read on.
The next avenue of growth for BEL is plant expansion at St. Ambroise. To this point we’ve said that St. Ambroise is a 100,000 ton plant. In truth 100,000 tons is the legal limit imposed by the provincial government. The assumption is that a more productive plant is more prone to environmental accidents, so BEL must prove that St. Ambroise can operate safely at a higher rate of throughput. St. Ambroise’s physical capacity is 140,000 tons. BEL has applied for a capacity increase from 12.5 tons/hour to 13.7 tons/hour (the plants are run 24 x 7). Management expects this to be approved without trouble in July. BEL has already done test burns to prove their capability to run at 13.7 tons, and the application is pending. If the incremental capacity is 90% utilized, this would add an additional $3.3 million in after-tax cash flow per year:

Rev./Ton $650
Incr. Tons 9,500
Cost/Ton $120 (incremental)
After-Tax CF $3.3 mill (taxed at 34%)

The truly spectacular growth opportunity for BEL is through the addition of new plants. To this end a second plant (Belledune) is currently 80% constructed. Belledune lacks an operating permit at this point, and is in the process of getting one. The public hearing process will be complete this month, compliance tests will be conducted by the end of August and full-scale production should begin in October, according to management. Management gives Belledune a 95% probability of going through. Essentially, short-hand you can say that an additional 100,000 ton facility (130,000 ton physical capacity, 100,000 ton permitted capacity) will roughly double free cash flow. But, there are efficiencies inherent in operating multiple plants, so long-hand gets you this:

(millions) St. A Belledune
Rev./Ton $650 $650
Capacity 108.0 90.0
---------- ----- -----
Soil Rev.(1) 70.2 58.5

VC/Unit 120 130
Var. Costs 13.0 11.7
Fixed Costs 10.0 10.0
----------- ---- ----
Oper. Costs 23.0 21.7

SG&A 8.0 1.0
Deprec.(2) 1.8 1.8
Oper. Inc. 37.4 34.0
Taxes (34%) 12.7 11.6
Unlev. FCF 24.7 22.4

(1) We’re only concentrating on soil processing revenues and costs in our model for simplicity’s sake. Gross revenue as-reported consists of four components: soil processing, project management, transportation and revenue from MRR (acquired Oct. 2002). In cases where transportation is included in gross revenue, it is always zero margin and thus irrelevant for our purposes. MRR did $4 mill of revenue in 2003 at an estimated 10% operating margin, so this is immaterial. Project management is a strange cat. In projects where BEL acts as prime contractor as opposed to sub-contractor (to-date Saglek is the only such project) BEL divvies up revenue between soil processing and site preparation/pre-excavation activities. Management tells us that the project management component of Saglek is 80% margin and the soil component is 50% margin, but in the end the entire project has similar 60% margins to sub-contracting work. This jives if you give the $10 million of project management revenue in 2003 (essentially all of the project management revenue associated with Saglek) an 80% margin and the remaining $20 million of the Saglek project a 50% margin, we arrive at a blended 60% margin. As a side note, this phenomenon also explains why Q1 sucked wind – essentially all of the project management revenue had been recognized in Q2-Q4 2003, leaving only low-margin soil processing revenue in 2004.

(2) We’re assuming maintenance capex on the plants will approximate depreciation, thus unlevered net income approximates unlevered free cash flow.

Note that BEL will spend $25 mill to build Belledune – this equates to a roughly one-year payback on the plant (100% ROIC), according to our model.
Management also expects a third plant to come online in 2006. Be it Kirkland Lake (a 200,000 ton plant which has been in the works since 1999) or a new site, management sounds confident that a third plant is a very real possibility, but for conservatism’s sake, given the difficulties involved in permitting new incinerators, we’ll leave that out of this model.

So we’ve got a fabulous business selling for 6x earnings power of the combined plants. Here’s our stab at a fair value for BEL:

Unlevered FCF 47.1
FCF Multiple 13x
Ent. Value 612.3
Cash (3/31) 23.2
Bell. Capex(3) (10.0)
Excess Rec.(4) 19.0
Equity Value 644.5
Diluted Shares 19.0
Stock Price C$34
Stock Price U.S.$25

(3) Management expects to spend $10 million in 2004 to complete Belledune, having already spent about $15 million.
(4) At 3/31/04 there was $19 million of excess receivables associated with Saglek. Management explains that the project was originally planned to take two years but that completion will take less than that, but the Department of Defense was budgeted to make payments over two years, making collection slow. The full $19 million should be collected by the end of Q3.

A 110% premium to today’s stock price – and if everything goes as planned it should all unfold within the next year or two.
So what’s our downside? Obviously, if none of these scenarios pans out we’re left with the stream of cash flows we had last year. We believe a realistic worst-case scenario is that the St. Ambroise expansion is denied and the Belledune operating permit is denied. We still believe 90% utilization at St. Ambroise is very attainable.

(millions) St. A.
Rev./Ton $650
Capacity 90.0
---------- -----
Soil Rev.(1) 58.5

VC/Unit 120
Var. Costs 10.8
Fixed Costs 10.0
----------- ----
Oper. Costs 20.8

SG&A 8.0
Deprec.(2) 1.8
Oper. Inc. 27.9
Taxes (34%) 9.5
Unlev. FCF 18.4

Unlevered FCF 18.4
FCF Multiple 10x
Ent. Value 184.0
Cash (3/31) 23.2
Bell. Capex (10.0)
Excess Rec. 19.0
Equity Value 216.2
Diluted Shares 19.0
Stock Price C$11
Stock Price U.S.$8

So on the upside we’ve got $13 and on the downside we’ve got $4 – a reward-to-risk ratio of 3.3.

Catalyst

1) Achieving 90% capacity at St. Ambroise
2) Capacity expansion at St. Ambroise
3) New plant coming online in Q4