Tuesday, April 23, 2013

Rocky Mountain Dealerships (RME)

Summary 
After reviewing my holding in Rocky Mountain Dealerships (RME) I continue to hold my position but am no longer accumulating shares. I am willing to swap-out the position if I find a better value opportunity, but I am not at this point liquidating shares to move to cash.

Background 
I originally purchased RME 2 ½ years ago and since then it has been a good (but not great) holding for me. However, after Titan Machinery’s (TITN) bloodbath recently, and after seeing a number of analysts upgrading RME in recent months (which frankly scares me), I thought now would be a good time to step back and review my thesis on the stock, its valuation, and what could go wrong.

Company History 
RME is an agricultural and construction equipment dealer. It primarily sells Case IH, Case Construction, and New Holland brands (as well as a number of “short” lines) through its 39 locations across Alberta, Saskatchewan and Manitoba.

Back in 2006, Matt Campbell (CEO) and Derek Stimson (President) merged their respective Hammer Equipment (primarily construction) and Hi-Way Service (primarily ag equipment) into one company and IPO’ed the firm in late 2007. Since then, the company has been on an acquisition spree, purchasing 15 independent dealerships since 2008. In 2012, sales broke down as (from 2012 annual report):

The strategy behind RME is beautifully simple. A larger dealer network can command better operating efficiencies and cost management than smaller, independent (“Mom & Pop”) dealerships. Management has noted that their target performance is a 15% (maybe up to 17%) gross margin with ~10% SG&A expense netting a 5% operating margin. This compares to small dealerships which typically have ~11% gross margins and an 8-9% SG&A overhead for a 2.5% operating margin. 

Now I’m usually very skeptical of economies of scale arguments, and this alone wouldn’t have gotten me to invest in RME. However, the economy of scale argument with a clear path to market share growth was more compelling. With its ability to dramatically improve margins, RME has been able to aggressively acquire its competitors, while still keeping acquisitions accretive to earnings.

Most importantly (in my mind), RME’s primary supplier (Case New Holland) supports a roll-up of small dealerships. Comments from equipment executives show that the big manufacturers (Case, Deere, Cat, etc.) want larger dealer networks for their own efficiency and operational reasons. When you have a better operational structure than the competition (higher margins), a deep pool for market share growth (acquisitions), and a franchisor (who effectively controls market share) in favor of a dealership roll-up, that’s a powerful business model.

All of this led to 2012 being a banner year for RME. Same-store sales were up 16.3% on the year. The company posted record EPS of $1.28, up 24% year-over-year. On top of this, there was a fair sized debt conversion expense in the year. If I try to adjust for this, I get “normalized” EPS of $1.45 (up 41% yoy).  Suffice to say, the ag economy on the Canadian prairies is booming ($8.00 wheat creates a nice cash flow). But are these types of EPS numbers sustainable?

Risks/Concerns 
The problem for RME is that acquisitions have fueled their growth, and those acquisitions are getting harder to come by. A quick search of the Case IH website reveals 27 ag dealerships listed in Alberta, and 18 of them belong to RME (67%). Saskatchewan may be an opportunity with 25 dealers listed and RME accounting for only 5 (20%), but there are other fairly entrenched dealerships in the province to compete with. Manitoba is similar to Albert, with 12 dealerships listed and RME accounting for 7 (58%). The company could jump south of the border into the U.S. for growth, but there is intense competition there from companies like TITN. There may also be room for RME to grow its New Holland presence, but that remains to be seen. 

Beyond growth, there is the always present worry that a downturn in the ag economy will hurt the company. We can argue about long-term ag environment and Malthusian trends, but that’s not the road I want to take today. What concerns me about a slowdown in ag spending is the following slide that RME management has often used regarding combine sales and the lifecycle of a combine: 


Combines are at the top end of the agricultural equipment food chain. New machines are running at or near $500,000 depending on options – in other words a major capital outlay. 

As the industry shifted away from leasing machines a few years ago (because that put the risk back on the equipment manufacturers [i.e. Deere, Case, Cat] balance sheet), dealers devised this ingenious lifecycle pyramid. The problem with pyramids, if my memory of stacking blocks in pre-school serves me right, is that it’s not the top of the pyramid that is important, but rather the base. 

I pulled some of the data from the above slide to generate the following table. Some of the estimates are my own / backed into to make the math work. 


What we see is that there are ~3,000 ~2.1 yr old machines sold annually (in Canada). There is demand for ~3,500 of those machines (from midsized farms). Basically, the slide is saying that there is a market imbalance. As an aside - I see this demand discrepancy as responsible for allowing the corporate farms to buy new machines annually, run them for a year (on warranty), and then sell them 12-24 months later without seeing much depreciation in value (strong demand from midsized farms). Basically the corporate farms are profiting on the backs of the small farmers (wow . . . doesn’t that sentence sound very Marxist). But . . . if this is happening, why aren’t midsized farms jumping up and buying new machines? Is this what happened in 2012 – boosting RME’s EPS? 

How real the top-end demand discrepancy is (the market clears doesn’t it) isn’t what concerns me the most – what concerns me is the bottom end of the pyramid. Play with the numbers even slightly (change the avg. years in use of machines for small farms from 7.6 to 8, 9, or 10) and the demand drop for combines reverberates up the pyramid (i.e. at 10 years, the “small” farms only demand 2,675 machines per year). As this manifests up the pyramid it would be ugly for farmers and devastating for RME. I have a tough time quantifying how likely this risk is, but common sense would argue that a decline in the ag economy (whether macro, price driven, or even short-term weather driven) would bring on a lengthening in the life cycle of a machine (yes, I know, the lifecycle for a combine is not 13.4 years – you can buy a 2000 model machine for ~$80,000 right now – not $0, but the theory/math holds no matter how you slice the lifecycle). 

Also, rival TITN machinery (based just across the U.S. border in Fargo, ND, and with 106 U.S. based dealerships) showed us recently what happens when the roll-up strategy stalls and margins contract (Midwest drought). The risk of margin contraction is inherent in the tight working capital situation of dealerships. Financing inventory on a floor plan is great when the business is booming, but floor plans add insult to injury when inventory turns slow and the bankers start demanding payments or additional collateral to maintain the line of credit. 

All of this leaves me asking what will drive growth in the future? As acquisitions dry up, market share will be hard to grow, and without market share growth, overall sales shouldn’t grow at much more than GDP type rates. 

Valuation 
The rub with RME is that it is cheap. Whereas TITN whiffed its earnings and investors are punishing it by rerating its P/E down from 14-15x to the current 11x, RME only trades at 9.3x my normalized $1.45 2012 EPS (adjusted for debt conversion expense, new long-term debt, 28% tax rate, etc.). If we look in terms of EV/EBITDA (adjusted for the debt conversion in 2012), RME trades at just over 3.0x. Of course, this assumes that the company’s floor plan is a payable (not included in EV) and not debt (included in EV). If I include the floor plan in EV, the ratio jumps to 9.5x. 

Free-cash-flow is also strong. If we assume depreciation/amortization and capex offset one another, than FCF is equal to net income. Last year an adjusted $1.45 in EPS would equate to a FCF yield of 10.7% - not too shabby. 

Catalyst 
I have a tough time coming up with readily achievable catalysts for RME: 
  • Further acquisitions are an opportunity, but the low hanging fruit has already been harvested. Growth into the New Holland line or geographic expansion might surprise me, but these come with new risks and the possibility of cannibalization. 
  • Margin expansion may be an opportunity. Management argues that 5-6 years after an acquisition the company’s gross margin should increase from ~15% to ~17% on a “larger installed equipment base” driving an “increase in high-margin product support revenue.” However, like acquisitions, the low hanging fruit is already gone and a 15% gross margin is consistent with what peers earn.
  • Long-term growth in the ag sector is an arguable catalyst, but with 16.3% same-store sales growth last year and my own worries about things like the combine lifecycle identified above, the industry may have pre-run this catalyst.
  • Free-cash-flow is the most obvious catalyst, with management conservatively being able to throw off $1.00 - $1.50 a year in FCF (7.5% - 11% yield), but this is somewhat offset by the risk of a low margin business in a cyclical industry using a floorplan payable for a large part of its financing. 
  • Value is often ridiculed as its own catalyst (cheap can get cheaper), but it’s hard not to argue that RME is a better value than peers. 

Updated Thesis 
My thought process is all over the place in this post – you can see me blindly stumbling around looking for an answer on this one. It just goes to show you that the sell decision is often much harder than the initial buy decision.

Although I’m no longer an accumulator of RME shares, I am not at the point of liquidation either. For my portfolio, I will replace RME shares with better opportunities as they present themselves, but I won’t sell RME to move to cash (yet). Maybe that’s just a cop out (a “hold” recommendation . . . really . . . I’m disappointed in myself), but I continue to view management’s ownership (~22%) and the value as weighted against the risks of slowing or declining sales, weakness in the ag economy, and the overall tight margins of the business.

Disclosure:  Long RME


Harvest Investor © 2013. All rights reserved. The content and ideas contained in this blog represents only the opinions of the author. The content in no way constitutes investment advices, and should never be relied on in making an investment decision, ever. No content shall be construed as a recommendation to buy or sell any security or financial instrument, or to participate in any particular trading or investment strategy. The author may hold positions in the securities and companies mentioned on this site. Any position disclosed on this site may be modified or reversed without notice to you. The content herein is intended solely for the entertainment of the reader, and the author.

Friday, April 5, 2013

Eagle Bancorp Montana, Inc. (EBMT)

Eagle Bancorp (EBMT) is the bank holding company for American Federal Savings Bank, a community bank based in Helena, MT. The company has 13 branches, primarily along the I-90 corridor in southern MT. As of Dec. 31, 2012, the company had ~$508 million in total footings and a market cap of roughly $40 million. 

EBMT has twice been written up on Value Investor’s Club, and the analysis put together by the two authors shames my feeble attempt, so I urge you to review those write-ups (both are available to “Guests” who register for a password). The most recent write-up from February 2013 does a nice job of explaining what’s happened recently with EBMT, specifically the acquisition of 7 branches from Sterling Financial in July 2012.

To recap, EBMT purchased: 

  • 7 Branches (Missoula [2]; Hamilton, Bozeman, Livingston, Big Timber, Billings) 
  • These complement the bank’s existing 6 branches (Helena [3], Townsend, Butte, Bozeman)
  • $44 million in performing loans. 
  • $182 million in deposits. 

Management paid a 4.3% deposit premium for the acquisition (EBMT paid Sterling $7.92 million for $182 million in deposits). The acquisition put to work excess capital that was raised with the second-step conversion back in 2010 (Equity/Assets dropped from 16.4% in June 2012 to 10.5% as of Dec. 31, 2012).

Beyond the branch acquisitions, another notable point regarding EBMT is its anniversary. On April 5, 2010, the bank completed its second-step conversion from a mutual holding company to a fully publicly-owned stock holding company structure. Without getting into the fascinating world of bank regulations (ZZZZzzzzzzzzzz), we can sum up to say that the OCC (savings bank regulator) severely limits a mutual company’s ability to be acquired in the three years following conversion. As of April 5, 2013 (what a coincidence – that's today!) those restrictions are eased (I say eased because this is still a bank, and the Federal Reserve Board (holding company regulator) / OCC (bank regulator) would still need to sign off on any acquisition). 

Now, my 2 cents: 

  1. The Sterling acquisition is not the slam dunk it appears at first blush. The cost of the deposits acquired ($182 million) was 0.72%. Doesn't sound like much does it? Yet, ~57% of those deposits were money markets, DDA, NOW, and savings. Those should be yielding ~0%. The ~43% in CDs would then have an implied yield of 1.67% - not exactly cheap money (I know, I know, 1.67% is cheap money in a normal environment, but the Fed’s ZIRP is anything but normal). Consider that a quick internet search reveals that I have to go out 5 years to get in the 1.50 – 1.75% on a CD right now. The reason for the heightened cost of acquired deposits is due to Sterling Financials problems in the financial crisis. Sterling was literally trying to avoid a run on its banks, and jacked up deposit rates to avoid insolvency (go look at its stock chart). Although I think EBMT can hold on to many of these deposits, it might not be perfectly smooth/easy/cheap.
  2. Although I like the idea of a potential acquisition catalyst, I’m not sure how high a probability we should assign such an outcome. From conversations with Montana bankers, there isn't a good in-state acquirer. Most of the other local institutions already have significant footprint overlap with EBMT. On top of this, the Sterling branches were shopped to everyone. EBMT didn't so much score a coup by grabbing them, but rather they were one of only a handful even interested in the branches (which makes the 4% deposit premium look a little expensive in my mind). Still, I'm not much good at identifying potential acquirers beforehand. Perhaps a mid-sized out of state institution wants into Montana, but I'm not building an investment thesis on it. 
  3. EBMT held up remarkably well in the financial crisis. I love a bank with solid underwriting. Just look at the ratios: 


Jun 05
Jun 06
Jun 07
Jun 08
Jun 09
Jun 10
Jun 11
Jun 12
Dec 12
NPL/Loans
0.47%
0.32%
0.13%
0.02%
0.75%
1.65%
1.57%
1.83%
0.70%
ALLL/Loans
0.53%
0.38%
0.33%
0.31%
0.31%
0.64%
0.96%
0.93%
0.85%
ALLL/NPL
114.4%
116.6%
244.3%
1641%
41.9%
32.1%
44.0%
29.1%
121.4%
NPL = Nonperformg Loans; ALLL = Allowance for loan and lease losses

Close (geographic) peers like GBIC and FIBK saw NPL/Loans spike to 4.5% to 7.8% in the 2009-2011 timeframe. 

Now some people are criticizing EBMT because they won’t be able to juice earnings with reserve releases as NPLs come down and the economy improves (FIBK/GBCI are now running ALLL/Loans at 2.5% - 3.85%, with ALLL/NPL of 100% - 130%). Although technically true, that’s just an accounting game. Games can influence short-term investors (a catalyst), but over the long-haul, I prefer the bank that never had to jump on the reserve/release roller coaster.

The stock looks moderately cheap if we think a number of things can go right:

  • (A) Leverage improves. As noted above, EBMT has now deployed some of its excess capital. This should help improve ROE levels by up to 50% (Assets to Equity improvement from 6.1x to 9.5x = 56% improvement). 
  • (B) The potential to write-new loans. Right now, EBMT is under-loaned at a loan/deposit ratio of 52%. Historically, the bank has run at ~75% loan/deposits. If they jumped up to a 75% ratio, this would be $97 million in new loans. I do not want them to do this quickly (I worry about underwriting standards), but even assuming a 5-10% increase in new loans annually that’s a $450,000 benefit to the bottom line and a 9bps boost to ROA annually (assumes $15 million in new loans annually, a 3% margin on new loans [consistent with most recent 10Q loan yield of 5.4% less securities yield give-up of 1.4% and a 1% hit for oversight costs = 3% yield increase] divided by $508 million in footings). 
  • (C) Now the question of what is a normalized ROA confronts us. Over the last 10 years, the bank has averaged 0.83%. Over that same time, the bank has posted an average efficiency ratio of 72%. This is inexcusable – peers (no, not management cherry picked peers, but rather banks right across the street from EBMT) are sub 60%, with a number below 50%. Whether through its new larger size, or just better control of the purse strings, EBMT needs to get down to 60% (as a side note, this is where I think an acquirer might salivate – it should be feasible to buy EBMT and quickly improve the profitability). A drop from 72% to 60% in the efficiency ratio would foster a 40% increase in ROA. Using the 10 year average 0.83% ROA, a 40% increase would indicate an ROA of 1.16%. Let’s use 1.0% to be a little more conservative. 
  • (D) Let’s put it all together. Now points B & C overlap somewhat (historically, EBMT has run at a loan/deposit ratio of 75%, so I can’t assume loan growth will increase ROA above the 10 year average). However, just looking at point A, which is an Asset/Equity ratio of 9.5x, and point C, which is an implied ROA of 1.0%, we get an ROE of 9.5%. If we want a 10% return on our investment, we can afford to pay 0.95x book value. At 0.8x right now, we’re buying EBMT at a near 20% discount.
  • (E) If we look back at the good old days before the financial crisis (2002-2008), EBMT had an average ROA of 0.89%. A 40% efficiency ratio improvement would get this to 1.25%, ROE to 11.88%, and an implied P/B ratio of 1.2x, which is 50% above current. 

Bottomline, I think EBMT is undervalued trading at 0.8x book value. It’s not incredibly cheap on an absolute / right now basis, but it has solid long-term potential, as well as a number of catalysts (acquisition integration, efficiency ratio improvement, improving economy, etc.) that should help it improve operations. Most importantly, management has historically been a top notch underwriter, which provides downside protection (margin of safety).


The importance of solid underwriting brings to mind an old J.P. Morgan quote. In December, 1912 Morgan testified before congress and was asked how he decided whether to make a loan. He replied, “The first thing is character.” One of the esteemed congressman suggested that perhaps factors like collateral might be more important, but Morgan retorted, “A man I do not trust could not get money from me on all the bonds in Christendom.” 

Disclosure: Long EBMT

Harvest Investor © 2013. All rights reserved. The content and ideas contained in this blog represents only the opinions of the author. The content in no way constitutes investment advices, and should never be relied on in making an investment decision, ever. No content shall be construed as a recommendation to buy or sell any security or financial instrument, or to participate in any particular trading or investment strategy. The author may hold positions in the securities and companies mentioned on this site. Any position disclosed on this site may be modified or reversed without notice to you. The content herein is intended solely for the entertainment of the reader, and the author.