Hudson Technologies (NASDAQ:HDSN) is a distributor and recycler of refrigerant gases. The company is poised to benefit from an EPA mandated phase out of virgin gas production of the leading refrigerant that will boost market pricing and volumes of reclaimed refrigerant where Hudson is the dominant player. Hudson stock remains depressed due to investor fatigue with the slow to develop shift in market dynamics and a general lack of Street recognition. The stock trades at 2-3X projected earnings when the market shifts, while the EPA is set to finalize substantially reduced virgin gas production quotas in the coming months. In the meantime, the company has expanded market share and is substantially profitable.
HDSN operates three business lines: aftermarket refrigerant distribution, reclaimed refrigerant sales, and refrigerant services. The only breakout by segment that HDSN provides in their filings is refrigerant sales (combined distribution and reclamation) versus service revenues. In the trailing 12 months refrigerant sales contributed 90% of revenues or $40.1 million, with service revenues at 10% or $4.5 million. Reclamation is currently a small piece of refrigerant revenues, probably somewhere in the $7 million range.
Aftermarket Refrigerant Distribution
Refrigerant gases are used in air conditioning and refrigeration units. They need to be replaced periodically due to leaking or gas loss during servicing. Hudson buys refrigerant, mainly from the big three producers (DuPont, Honeywell, and Arkema, who control approximately 90% of the virgin refrigerant market) and distributes it to contractors, large end users, and other distributors for aftermarket use.
Management targets gross margins of 30% in this segment and they have generally ranged from 20-30%. The refrigerant aftermarket is driven by demand for cooling (primarily air conditioning) and is generally a low single digit grower although it does have some exposure to weather and the economy. Cool summers reduce demand and in a weak economy customers can temporarily delay unit maintenance. HDSN saw refrigerant revenues drop in '09 due to customer inventory destocking at the onset of the recession, but refrigerant segment revenues have grown double digit percentages every other year since 2005:
Although the company does not explicitly break out new versus reclaimed refrigerant sales, the sense from management is that revenue growth has come primarily on the back of increased market share on the distribution side. The overall refrigerant market has been flat and reclaimed sales have been perennially weak.
Hudson is the largest refrigerant reclaimer in the U.S. with two adjacent facilities in Champaign, Ill. and a management estimated 20% market share. The business model here is that HDSN will pay the end user or distributor for their used refrigerant gas and bring it back to their reclamation facility. Contaminants are separated out of the gas, the gas is laboratory tested to insure purity, and the gas is repackaged and sold at the market price for virgin gas. Management targets 50% or higher gross margins on reclaimed gas, which they think they can easily hit in a more robust environment for reclamation.
While there are several smaller players in the reclamation industry, HDSN is the industry leader both in terms of production capacity and sophistication of their process. The major refrigerant producers left the reclamation market years ago and do not appear to have a desire to get back into it, probably due to its small size relative to the virgin gas market.
While there are not huge barriers to entry in the reclamation business, HDSN does have some noteworthy competitive advantages. HDSN has strong relationships with the major gas producers. HDSN serves as the exclusive national reclaimer for DuPont and Arkema and has smaller deals with Honeywell distributors. These relationships allow HDSN to utilize the producers’ extensive distribution networks to ensure a supply of used gas for reclamation. The gas producers’ local distributors serve as collection points for end users to drop off dirty gas for reclamation.
Large scale reclamation technology is proprietary. The EPA (Environmental Protection Agency) commissioned a 2010 report on reclamation practices, which noted the vast difference between the systems of the smaller and larger reclaimers in terms of quality and efficiency. HDSN has the largest reclamation capacity and is widely regarded in the industry as having the most advanced reclamation technology.
HDSN provides refrigeration diagnostics and preventative maintenance, as well as energy optimization solutions. The company has attempted to position their client-site services as their growth engine for over a decade, but have not had much success. Service revenues have ranged from $3.5-4.6 million since 2005, although they are on the upswing in the past year. HDSN does not disclose gross margins for the segment, but management gives the sense that they are significantly higher than on the distribution side.
The heart of the investment thesis in HDSN surrounds the phaseout of HCFC refrigerant gases, particularly R-22, which is being mandated by the EPA. R-22 is currently the most widely used refrigerant and the primary target of the EPA phaseout. The Montreal Protocol is an international agreement, to which the US is a party, that phases out the production of various gases damaging to the ozone layer. The phaseout of CFC gases (of which R-12- branded as Freon- was the most popular) has already been completed. HCFCs such as R-22 are less damaging to the ozone than CFCs, but they too are being phased out per the next steps in the Protocol.
The Protocol calls for a gradual reduction in HCFC production from a baseline level leading to an eventual total ban on production. The EPA has chosen to accelerate that schedule to target a total ban on R-22 production by 2020. They are implementing the phaseout by allowing a declining percentage of R-22 demand to be met by virgin production. As virgin gas production is phased out the EPA wants to increase the usage of reclaimed gas to make up the gap in demand. Reclaimed gas is not subject to the limitations of the Montreal Protocol.
However, R-22 prices must be high enough for reclaimers to be able to offer gas users enough of an incentive to bother to return the gas and not just vent it into the air. While venting of R-22 is technically illegal, the EPA cannot enforce the rule and venting is a widespread practice with the EPA estimating that over 90% of dirty gas is currently vented. The EPA’s plan is that limited virgin gas supply due to the declining allocations will cause R-22 prices to rise and spur reclaimed gas to fill in the demand gap.
In the EPA’s target scenario HDSN would benefit in several ways:
Higher R-22 prices would equate to more gross margin dollars on the distribution side (even assuming gross margin as a percentage remains a consistent spread between what HDSN pays producers and what end users pay HDSN).Higher reclaimed gas sales volumes would boost reclaim revenues.Higher R-22 prices would boost reclaim gross margin percentage as less of the selling price would need to be given to the end user to buy the gas back.
What’s Happened So Far?
The EPA started the more aggressive phaseout of R-22 with the 2009 Final Rule that went into effect in January 2010. Firstly, the EPA banned the sale of any new cooling equipment that utilized R-22 to eliminate the creation of any new R-22 demand. (Although there is a loophole to that rule that has been exploited – more on that later.) Secondly, the EPA issued new R-22 consumption (defined as production+import-export) allocations by gas producer for 2010 to 2014. The EPA attempted to estimate annual R-22 demand for that period and assigned a declining percentage of demand to virgin gas production with the theory that reclaimed gas sales would fill in the demand. Reclaimed gas sales would stay at a constant absolute level throughout the period, but rise annually as a percentage of total consumption and ensure a smooth transfer to the total ban on R-22 virgin production (chart from EPA’s Dec 2009 “The US Phaseout of HCFCs” report):
In 2000, R-22 sold for around $1 a pound. From 2006-2008 the major producers increased pricing steadily in anticipation of the phaseout and R-22 hit a peak of $5 a pound in the middle of 2008. Reclaimed R-22 also rose from 1,960 metric tons in 2001 to 4,556 metric tons in 2008 (based on EPA numbers- which should be very accurate as all reclaimers have to report their production to the EPA). Yet since 2008, R-22 pricing and reclaimed volumes have both fallen despite the implementation of the EPA phaseout program in 2010. Current R-22 pricing is $4 a pound or a 20% drop from the 2008 peak. Reclaim volume in 2010 was 3,584 metric tons, or 21% lower than 2008, and based on the comments of reclaimers to the EPA in late 2011 they expect 2011 volumes to be about the same as 2010.
Clearly, the EPA’s plan to raise R-22 pricing and hike reclaimed gas volumes has not worked thus far. A court order in August 2010 provided an opportunity for the EPA to revisit the phaseout process. In brief, two gas producers claimed the EPA miscalculated the percentage of R-22 production allocations they should have received for the 2010-2014 period. The producers won the court case, and the court order vacated the EPA’s production allocations in the 2009 Final Rule. While the EPA put in a stop-gap interim rule for 2011 to incorporate increased allocations to the two snubbed producers, the EPA also took comments on how to revise the phaseout process for 2012 to 2014 to better meet their goals of higher R-22 pricing and increased reclaimed gas production.
The comments to the EPA of both refrigerant producers and reclaimers submitted in September of 2011 unanimously agreed that there is an oversupply of virgin R-22 in the market, which is causing the depressed pricing. The stakeholders raised several theories as to the cause of oversupply:
The EPA’s demand study was done prior to the onset of the recession and the declining economy contributed to lower R-22 demand.Due to the advanced warning the industry received the transition to non R-22 cooling units has been implemented more quickly than anticipated, thereby reducing R-22 demand.Distributors and end users built R-22 stockpiles prior to the initiation of the ramped up production phaseout in 2010.The EPA has initiated a program called “Greenchill” to partner with large food retailers to reduce the environmental impact of their refrigeration usage. The success of the program reduced aftermarket R-22 demand on the part of food retailers, which is a major segment of R-22 usage.[/list]On January 4, the EPA released their proposed rule for revised R-22 production allocations for 2012-2014. Incorporating stakeholder comments and interviews, the EPA proposed a range of production allocations, which would result in an 11 to 47 percent reduction versus the original allocations. The EPA proposed adjustments to the original allocations based on three new or revised assumptions:
- 6,000 mt of demand to be met by surplus inventory drawdown
- Between 12,500 mt and 19,700 mt of demand to be met by reclaimed volume (original allocation assumed 12,500 mt)
- Between 1,600 mt and 5,500 mt of demand met by R-22 reused within large food stores (impact of Greenchill program)
This table from the EPA Adjustment memo of 12/16/11 bridges the estimated demand to the virgin gas allocations proposed in the new rule:
This chart from the EPA memo compares the proposed allocations to the original allocations in the 2009 Final Rule:
The EPA is taking comments for 30 days from stakeholders and will issue a final ruling thereafter. DuPont requested a 20% reduction to the original allocations in a letter to the EPA before the release of the proposed rule. While the producers are in favor of reduced R-22 production to raise R-22 pricing and shift demand to replacement gases like R-410a, they also have to balance meeting the short-term needs of their customers, which is why DuPont did not ask for a sharper reduction. The reclaimers will surely ask for the maximum proposed reduction in their comments.
There are several questions remaining whose answer will impact the R-22 phaseout going forward.
Where will the EPA’s final ruling fall on the proposed range? The proposed range of reductions is quite broad (11-47%). But even in the worst case scenario we will see an 11% reduction to the original allocation, which was already declining annually. Based on DuPont’s comment if we assume the producers will ask from something in the middle of the range and the reclaimers will ask for the max reduction then the final rule should end up on the higher end of the reduction range.
Another unknown is when the EPA will issue their final rule. The peak selling period for R-22 is the first half of the year prior to the summer cooling season. So the earlier the EPA releases the reduced allocations the more of an impact the proposed rule will have on production and pricing in 2012. The production of R-22 is technically illegal without R-22 allocations, as has been the case since January 1. However, if the EPA expects a significant delay in the release of the final rule they will probably release a “no enforcement” letter that will allow producers to operate based on the original allocations until the final rule is released. (The EPA released such a letter last year after the court order vacated the 2009 Final Rule allocations.) Given that there probably will be nearly unanimous sentiment expressed in the stakeholder comments on the proposed rule I do not expect a significant delay in the release of the final rule.
The biggest question is whether the reduced production allocations will be low enough to raise R-22 pricing. One potential problem is that R-22 inventories are not officially reported to the EPA, and thus the EPA is working off of inventory estimates derived from anecdotal evidence from stakeholders. Another issue is that the EPA did not fundamentally revisit their model for estimating R-22 demand, and in fact revised the estimates for R-22 demand about 6% higher in the proposed rule. The EPA did not go into detail as to why they revised their demand estimates higher.
My guess is that it is due to the unexpected growth in R-22 units due to the dry-ship phenomenon. While the sale of new cooling units charged with R-22 was banned in 2010, manufacturers noted that the rule only prohibited the sale of units actually shipped with R-22 gas. New units could be “dry-shipped” without R-22 and then charged with R-22 in the field. While the air conditioner manufacturer Carrier petitioned the EPA in February 2011 to close the loophole, it does not appear the EPA will act to do so. The EPA might be willing to overlook the dry-ship units in the short-term as it will increase demand for R-22 and help boost R-22 pricing. However, the EPA’s revised demand estimates may be taking this increased demand into account at least partially.
The Air Conditioning, Heating, and Refrigeration Institute (AHRI) estimates that 17 million residential R-22 units will be replaced from 2010-2015. If the vast majority of those are replaced with dry-shipped units as appears to be the case at present then that could drastically extend the long tail of R-22 demand with units expected to last for twelve to fifteen years. In their 2009 report, which assumed no new R-22 units after 2009, the EPA projected that 86 million R-22 AC units would in operation in 2015. 17 million dry-ship units would add 20% to that number.
Can we estimate R-22 demand ourselves? The best hard data points on R-22 demand come from 2010. The EPA has said that 86%, or 42,973 metric tons, of the 2010 consumption allocations were used by producers, and that reported reclaimed R-22 volume in 2010 was 3,584 mt. So in 2010 demand needed to be met by either new or reclaimed gas was 46,557 mt. I think we can assume that demand met by stockpiled inventory has at least not increased since 2010 when the more stringent production allocations went into effect.
The EPA estimated that demand would fall by 15% from 2010 to 2012. Taking 15% off of the 2010 number would yield 39,573 mt of demand for new or reclaimed gas in 2012. Subtracting the 12,500 mt the EPA is targeting on the low end for reclaimed gas would yield a 27,073 mt allocation to new gas. For 2012 the EPA has proposed 25,100 mt allocated to new gas on the low end and 36,200 mt on the high end. This is before taking into account dry-shipped units, which have certainly added some demand since 2010. Even taking the middle of the allocation range, or 30,650 mt, would mean 8,923 mt would need to be met by reclaimed demand. That is still more than double the 3,584 mt of reclaimed production in 2010 and would almost certainly mean R-22 pricing would need to be much higher.
Precedents: R-12 and Europe
There are two phaseout precedents we can look at to see how the R-22 phaseout might play out. One is the phaseout of CFCs in the 90s. The most popular CFC is R-12 or Freon, used mainly in car air conditioners. R-12 production was phased out from 1993 to 1996. This was done through a combination of EPA production limits and an excise tax on production. Over that time period the price of R-12 rose from $3.50 per pound to $17, and eventually rose to nearly $30. While the R-22 phaseout is being implemented across a longer time span and there is no excise tax proposed, R-22 is a much larger portion of the refrigerant market than R-12 was (60% vs. 20%).
The EU is well ahead of the U.S. in its R-22 phaseout, having completely banned virgin R-22 production in January 2010. Current R-22 pricing in Europe is approximately $15 per pound compared to $4 in the U.S.
Earnings Power and Valuation
I think HDSN can earn approximately $.50 of EPS in the near future with a potential peak of $1.00 or more as the R-22 phaseout is implemented. You can see my full model here. Here is the summary of base and upside scenarios:
Here are the details of my base case:
- My estimate above of ~40,000 mt for 2012 R-22 demand for new and reclaimed gas, the mid-range of the EPA’s proposed 2012 new gas allocation or 30,650 mt, which leaves 9,350 mt of demand to be met by reclaimed gas.
- Average R-22 price of $8. EPA memo 12/16/11- “Currently virgin HCFC-22 is priced at around $4.50 per pound; some industry representatives suggest that this price would have to exceed $8.00 per pound for reclaimed HCFC-22 to become competitive”. The 9,350 mt of reclaimed demand would be more than a doubling of the current reclamation market.
- HDSN reclaimed gas market share of 20%- HDSN management
- HDSN 3.3% market share for aftermarket refrigerants- based on my estimate of HDSN ~$33 million in TTM distribution revenue and $1 billion market size (HDSN management)
- TTM service revenues of $4.5 million
- HDSN target gross margins of 30% on distribution and 50% on reclaim, my estimate of 50% gross margins on service.
- Selling expense at 6% of revenues, in line with historical, assume scales as percentage of sales due to commissions.
- G&A expense of $4.5 million. HDSN has done a very good job of managing G&A expense as sales have grown. G&A was $3.2 million on $23.5 million in revenues in 2006 and $3.9 million on $44.6 million in revenues in the last twelve months. There should be a good deal of leverage on this line.
- TTM interest of $1 million
- Historical normalized tax rate of 38% (34% federal, 4% state)
- Fully diluted share count of 24.7 million[/list]This scenario produces $.46 of diluted EPS. Cash EPS will be somewhat higher as the company has $20 million of NOLs although they are subject to $1.3-2.5 million annual limitations.
As the allocation to virgin gas falls annually, the price of R-22 should continue to rise to meet the increased need for reclaim supply. So although total demand for R-22 will be falling, HDSN earnings should rise as a higher percentage of the demand is met with reclaimed gas, where HDSN has higher margins and much larger market share. In the upside scenario, which could play out in 2014:
- R-22 total demand falls 20% from my 2012 estimate to 32,000 mt, reclaimed gas reaches EPA target of 12,500 mt
- R-22 pricing at $15, still below precedent of R-12 phaseout and at current Europe R-22 levels[/list]With other assumptions remaining the same, in that scenario HDSN earns $1.02 in EPS.
If the R-22 phaseout plays out anywhere close to what is expected, HDSN is very cheap. R-22 reclaim demand will obviously not last forever so we should not assign a normal multiple to peak earnings. But the tail for R-22 demand is being rapidly extended as long as the EPA continues to allow R-22 dry shipped units. And as the virgin gas allocation continues to drop more of the market will be shifted to reclaimed gas where HDSN earns higher margins. This is clearly more than an opportunity for a single year of elevated earnings. To frame the big picture opportunity, the EPA estimated there will still be R-22 demand of 18,200 mt in 2020 by which time virgin gas production will be completely banned (and that does not take into account any demand from dry-shipped units). In 2011, total reclaim volume will probably come in around 3,600 mt. I think it is reasonable to put a 5X EPS multiple on $.74 of EPS, which is in the middle of the base and upside cases. That gives us a $3.70 stock. You can quibble on the multiple here, but in a nutshell if R-22 pricing moves up significantly then HDSN stock should do very well.
It also is worth noting that the replacement gas for R-22, R-410a, causes global warming and there have been motions to put an R-410a phaseout into the Montreal Protocol as well as a bill proposed in the House of Representatives. There is a chance that by the time the R-22 phaseout is done the reclaim volumes will be replaced by R-410a.
There is limited downside to a scenario where somehow the R-22 phaseout thesis does not play out at all. If we normalize trailing gross margin from 21% to 25% to smooth out the impact of FIFO accounting then HDSN would have done $5 million in EBIT on trailing twelve month revenues of $44.6 million. A normal multiple of 8X EBIT gets you to an enterprise value of $40 million, which is about equal to the current enterprise value with the stock at $1.65.
A risk to the thesis may be that other reclaimers could be willing to accept something less than 50% gross margins by either 1. selling their gas for less or 2. paying out more to users to secure dirty gas supply.
- Pricing for reclaimed gas tends to be in line with virgin gas pricing, which itself is typically uniform across the major producers. Many reclaimers are also distributors who buy from the producers, so they are unlikely to want to undercut their key suppliers on price.
- Having to increase payments to collect the dirty gas from distributors is a legitimate risk. In the 2009 Final Rule the EPA stated that the reclaimers had combined capacity of 16,329 mt. So there is a huge amount of excess capacity currently with 2011 reclaim volumes estimated around 3,600 mt. And there would still be plenty of excess capacity even if the EPA target of 12,500 mt of reclaimed gas is hit. This might incent reclaimers to bid against each other for dirty gas. HDSN probably has some leg up here due to their relationships with the major producers’ distribution networks, although they could always go elsewhere for a better price. However, HDSN has the most capacity of any reclaimer, and although there is no hard data available they appear to be the largest by a sizable margin (the typical reclaimer has several hundred thousand pounds of capacity versus several million for HDSN). HDSN enjoys economies of scale that the smaller reclaimers do not possess and has a significantly lower cost structure at higher reclaim volumes. That should enable them to offer compensation to buy back gas equal to that of their smaller competitors and still enjoy high margins.
Capex and Working Capital
HDSN maintenance capex is very minimal at $400 to $600 thousand (about equivalent to depreciation). Capex in 2011 will probably be closer to $1 million due to the $500 thousand contribution to the European JV. Working capital is seasonal with inventory builds in the first half of the year, and will certainly be a use of cash as sales grow. But HDSN currently has $5 million in cash and a $15 million line of credit available with $4 million currently utilized so they should not have an issue financing future sales growth. Their use of the line of credit peaked at $8.8 million in the second quarter of 2011.
European Joint Venture
In July 2011, Hudson announced it would be initiating a joint venture in Europe with two European partners. HDSN will contribute $500,000 and own 40%. The JV will enter the refrigerant reclamation and energy optimization markets. As noted above, the R-22 phaseout in the EU is well ahead of that of the US with no new virgin production allowed and R-22 pricing at $15 a pound. Additionally, the European reclamation market is underdeveloped and HDSN thinks it can utilize its superior technology and know-how to build a major reclamation business in Europe together with their partners’ distribution networks. HDSN thinks they will also have greater success in Europe selling their energy optimization services as the EU is also more advanced than the US in terms of offering incentives for carbon credits and other forms of energy usage reductions.
Management has not provided much guidance on the European venture other than that it will be operational by the end of 2011 and accretive to earnings in 2012. There appears to be little downside given the minimal capital contribution, and the JV could grow into a material contributor to earnings based on the reclamation dynamics in the EU. However, I have not accounted for any contribution from the JV in my earnings scenarios.
Why does the opportunity exist?
- I think the primary reason the stock is so cheap is investor fatigue with the R-22 phaseout story, which was supposed to have started to play out in 2010. The stock got up to $2.85 in early 2010 and has traded lower since then based on disappointing R-22 pricing and low reclamation volumes. Investors are growing frustrated waiting despite the company performing well given stagnant R-22 pricing and the recent positive developments with the EPA allocation reset.
- HDSN issued 2.7 million shares and 1.3 million warrants in July 2010 at $2 a unit (warrants exercisable at $2.60). This was obviously expensive capital and did not help the stock price. Management felt they needed some liquidity to prepare for the increased working capital demands with higher sales and were worried about getting shut out of the market. I don’t foresee a reason they should have to raise equity in the near future.
- The last two quarters earnings have been weak due to gross margin pressure with gross margins coming in at 18.4% in Q2 and 12.3% in Q3 compared to the usual 20-30%. Management has said the decline is due to using FIFO accounting in a falling R-22 price environment. Even if R-22 pricing remains flat gross margins should stabilize as they work through the higher priced inventory.
- There are the classic micro-cap factors as well- the stock trades less than 30 thousand shares a day and there is no sell-side coverage.
Insiders own 38.3% of the company, with the founder and CEO Kevin Zugibe owning 22.8%. I view the insider ownership as positive here given that executive compensation is relatively modest. Hedge funds Becker Drapkin and Marathon Capital own 9.1% and 7.2% respectively. Although Becker Drapkin has often gone activist, they have been passive in this investment. One unknown is what management will do once the company starts generating some excess cash, but given the insider ownership and hedge fund holdings I think there are good odds that shareholders will be fairly rewarded.
The timing on this investment is difficult to predict. Clearly R-22 pricing has to go higher at some point as production allocations fall annually and inventory stockpiles run out. The question is when that will happen. If the EPA gets the final rule out by the end of February and their revised assessment of supply and demand is accurate then R-22 pricing should start to rise by the second or third quarter. If they delay the ruling or get the supply/demand wrong again then it is hard to know how much longer it will take, but time is HDSN’s friend as the R-22 production allocations fall and stockpiles dwindle. The stock has had a nice run in the last few days in the wake of the EPA proposed rule, but still seems cheap if the R-22 thesis finally begins to play out. Even with the stock $1.60-$1.80 there should not be much long-term downside given the current normalized run rate EBIT of $5 million.
Disclosure: I own shares of HDSN.
Elie Rosenberg runs a value investing research website at valueslant.com. Sign up here to get his free value investing ideas and analysis by email and get his free ebook, "16 Ways to Find Undervalued Stocks."