Canterbury Park (CPHC): A Stock Selling for Less than the Sum of Two Parts – A Card Casino and 127-Acres of Land (Plus You Get a Horse Track for Free)
Canterbury Park (CPHC) is a sum of the parts stock.
After our experiences – and when I say “our”, I mean my decisions to buy – Maui Land & Pineapple, Keweenaw Land Association, and Nekkar – Andrew has a sticky note on his desk that says: “When thinking about SOTP, think STOP”.
Canterbury Park (CPHC) is a sum of the parts (SOTP) stock. Since we’re thinking “SOTP” should we also be thinking “STOP”?
Yes, Canterbury Park is a sum of the parts stock. But…
That doesn’t mean it is primarily an asset play. Though it might be. I’ll talk about the company’s horse track and card casino in a second. But, first let’s get the hardest part for me to value out of the way.
I find it difficult to value the real estate assets of this business. So, I will be judging them based in large part on the range of per acre transaction prices – for both sales of land and purchases of land – I found in the company’s filings. Amounts paid or received per acre seem to range from about $180,000 to $385,000. Some of these deals are a bit more complex – for example, it’s difficult to determine what the price per acre the company received was when it exchanged land for an equity stake in an apartment complex or something like that. In one specific example of this – I would say the company exchanged about 1 acre of land for every 8 apartment units (I mean here the equivalent of owning 100% of 8 units, actual ownership is a minority stake in a greater number of units). Well, what exactly are 8 apartment units worth in the area? I don’t know. Could 8 apartment units be worth something in that same $180,000 to $385,000 range? Could it be more? I’d have to do a lot more research – and be a lot better at understanding real estate investments – to get definitive answers to how much Canterbury Park’s real estate is worth. It’s an important part of the investment case here. But, it’s not one I can evaluate well.
The real estate not being used by the business is 127 acres. Total real estate ownership is more like 370 acres. But, most of it is tied up in the horse track – so, I’ll limit discussion to the 127 acres that is planned to be developed into apartments, townhomes, extended stay hotels, etc. The lowest values I found for actual transactions the company has engaged in were around $180,000 per acre. If we assume the company receives the equivalent of $180,000 in value – sometimes in cash from sales of land, sometimes from equity stakes in joint ventures that rent out apartments for years to come, etc. – we’d place a value of about $23 million on all this real estate. It might be worth $25 million.
It could be worth a lot more than that depending on how it’s developed. For example, the amount of infrastructure Canterbury Park expects to put into this development – and then be repaid by the city through the additions to tax revenue created by these infrastructure investments – is around the full amount I’m discussing as the value of all the land. I don’t know enough about real estate development projects to know if that makes sense. Would you put $20 million to $25 million into public infrastructure improvements in an area – and expect enough incremental tax revenue from the new development to eventually recoup those costs – and then only value the underlying acres at that same $20 to $25 million? I don’t know. This just isn’t an asset I understand well enough.
So, I’m going to arbitrarily assign all the developable land Canterbury Park owns a value of $25 million and just move on with my analysis. I should point out that – unlike some stocks I’ve looked at in the past with a lot of empty acres – Canterbury Park is really far along in the development process here. Their other operations produce cash flow. They have good access to credit. They have a partner lined up for the two phases of the planned apartments. It’ll be 300 units (phase one) plus 300 units (phase two) equals 600 units and CPHC will own about a quarter of the equity – so, that’s like owning 150 apartment units outright. And this is just one of the assets Canterbury Park can get out of all this development. So, I don’t doubt there could be a lot of upside here. I just am not going to assign a present value of more than $20 million to $25 million to the land they own today. I’ve decided – for the purposes of this write-up – to just use $25 million. There are about 4.6 million shares outstanding. That’s about $4.50 to $5.50 per share in yet to be developed real estate value. In other words, we can just take $5 off the stock’s current price of $12.49 a share and then check to see if the rest of the company’s businesses are worth more than $7.40 a share.
What are the company’s other businesses? Canterbury Park lists 3 other segments: horse racing, food and beverage, and the card casino. I’m only going to assign a positive value to the card casino. Let me explain why. The way Canterbury Park breaks out its segments allows us to see the segment level assets and earnings of each segment. We know food and beverage pays – this is within the company – a large part of its gross revenue (on days on which there are live horse races) to the horse racing segment. We also know that food and beverage’s segment assets as shown in the 10-K don’t include facilities it uses that are listed as either assets of the card casino (which are minimal) or the horse racing segment (which is a huge amount of the total assets of the business). Even with what I’ve discussed above – the return on assets at the horse racing segment is very low. First of all, we’re talking about segment level results – without certain corporate overhead applied to it. Secondly, a lot of the assets of the horse racing track have already been somewhat depreciated on the books. So, the segment level book value of the assets may be lower than the replacement cost of the assets if you had to build a new track today. Third, there’s what I’ll call “a lobbying alliance” between Canterbury Park and a local Indian tribe (that operates the nearby Mystic Lake Casino) which results in about $10 million per year of indirect benefit to Canterbury Park generally and primarily to Canterbury Park’s horse racing segment specifically. The Shakopee Mdewakanton Sioux pay over $7 million a year in purse enhancements that allow races at the horse track to pay out more to the winners than would be the case without these enhancements. Bigger purses mean a race track can attract both better quality horses and also simply more horses (“a bigger field”). Higher quality competition and larger fields increase the amount of betting on a race, the number of spectators in attendance, etc. Without this co-marketing agreement (as the company calls it – I really consider it a lobbying alliance) the live races would be more costly to run and/or produce less revenue from wagers and other sources. As a horse track hosting live races, Canterbury Park also can make money by “simulcasting” other races around the country. They get a meaningful (relative to the total amount of revenue brought in by all their horse racing activities) amount of revenue from bets made on these simulcasts. Even with counting all assets at their somewhat depreciated levels, including the benefits of the Shakopee purse enhancements, counting the simulcast revenue the same as the live race revenue, etc. I sometimes see returns in this segment as low as about a 4% EBITDA (and cap-ex requirements in this segment are very real – so, I think EBITDA flatters returns here) return on assets. Even if I include food and beverage returns as if they belong to horse racing – when, really some of that return is tied more to the card casino – I get no better than 8% EBITDA returns on that side of things. If horse racing was a separately traded entity – I don’t see how it could have reported after-tax earnings greater than 5% of its assets. I think returns would be lower than that. And unleveraged cash returns here could be quite a bit poorer than 5%. This segment doesn’t earn its cost of capital. Growth here won’t create value. And I have serious doubts about whether using the acreage they do for a horse racing track is anywhere near the highest and best use you could get for this land. But, it’s a necessary use – because of the next segment we’ll be talking about.
Canterbury Park has gaming licenses because it is a horse racing track. This kind of thing happens a lot where a location that was originally used for one kind of gambling is eventually allowed to conduct other kinds of gambling at the same site. Horse tracks originally ran live races spectators could bet on. Then they were allowed to televise other horse races around the country (year round, instead of only during their own limited race seasons) and take a cut of bets made on those races through their locations. And then – at least in the case of Canterbury Park – other forms of gambling beside horse racing were allowed. Here it is a card casino. Canterbury Park has permission to run an 80 table card casino. The company can run either banked (the house puts up its own money) or unbanked games. They choose to run only “unbanked” games. Winnings come from player pools – a lot like the way betting on a horse race works – and the house does not risk any of its own money. Technically, Canterbury Park does kind of use some of its own money in the sense that some of what it is allowed to take on games it then recycles back into potential winnings – often things like a “progressive jackpot” – as a sort of marketing enhancement. This isn’t really all that different than like player loyalty programs that casinos run which include cash vouchers. Canterbury Park has one of those too. If you look at the company’s balance sheet, it includes some of the stuff I’m talking about (the limited form of “banking” it does – which really is just a rebate / bonus type system) under “restricted cash”. The card games Canterbury Park runs are of two types. One is poker. This accounts for about one-third of the card casino’s revenue. The other is table games. The games we are talking about are things like: blackjack, baccarat, pai gow, etc.
The card casino is quite profitable. It uses minimal assets. It’s not a very big space. Segment level EBIT at the card casino averaged $7 million in 2017 and 2018. Some renovations were done and there was a regulatory change – both potentially a bit positive – between the time those results were reported and today. Also, if I am understanding Minnesota’s minimum wage law correctly, I believe all aspects of this company’s labor expenses will look better over the next 4 years than the past 4 years. I think Minnesota raised the minimum wage by 8% a year in each of the last 4 years and is likely to raise it no more than 2.5% per year for each of the next 4 years. This company employs a very large number of workers at minimum wage, seasonal workers, college students, part time workers, etc. So, expense control and customer service is likely to be easier over the next few years than the last few years. Anyway, we’ll just call Card Casino earnings before taxes $7 million. I’m going to assume a 30% tax rate here (federal taxes are 21% and Minnesota corporate taxes can be close to 10%) which may be overly conservative, but is a nice round number. That gives me an estimated after-tax earning power for the card casino of $4.9 million. There’s really no assets directly attributed to this segment, minimal depreciation, it’s a cash business, etc. So, I’m going to just round that up and assume this thing generates about $5 million a year in after-tax free cash flow.
What’s that worth?
Historically, stocks have often traded above 15 times P/E ratios. So, the card casino might be worth $5 million times 15 equals $75 million. The company has 4.63 million shares outstanding. So, $75 million divided by 4.63 million equals $16.20 a share. As I’m writing this, the stock is trading for $12.50 a share. And I just said the card casino might be worth $16.20 and the real estate might be worth another $5. So, you have a stock trading for less than $13 a share that might be worth more than $21 a share. Or, to put it another way – if the real estate is worth $25 million and the card casino is worth $75 million, then CPHC’s enterprise value should be $100 million while it’s actually just $60 million.
Are their risks?
Yes. The card casino appraisal I gave was just $75 million versus $60 million for the whole stock. That’s without corporate costs, etc. attached. If the real estate development doesn’t add value and this company dilutes shareholders at the rate it has in the past – shares outstanding have sometimes risen as fast as 2% a year, that’s a big drag on your future returns in this stock – this could easily be a mediocre investment.
New competition from other gambling in the area could be a problem. The agreement between the Shakopee (who operate Mystic Lake Casino just 4 miles from Canterbury Park) and Canterbury Park runs through 2022 and includes requirements for Canterbury Park not to expand into other gambling and to oppose the expansion of gambling elsewhere in Minnesota. But, this is always a risk with any gaming stock.
I didn’t go into much detail here – but, this company’s facilities are very different from those of Gamehost. The Alberta casinos Gamehost runs are pretty rough and rundown compared to the state Canterbury Park has kept its facilities in. I’ve looked at reviews of the casino and race track, what the company says about its own facilities, cap-ex spending, etc. and compared it to what I know about horse tracks in the U.S. and regional casinos. I think this company is maintaining its site very well by industry standards. There are many horse tracks in worse shape than this one. And the development around the track could be synergistic. When I say the track isn’t really worth assigning a value to – I don’t mean to suggest it isn’t bringing in traffic (attendance averages 6,500 people on the nearly 70 live racing days a year, for example). The renovation of the card casino included a change to provide a “grand entrance” from the live racing and event space to the card casino. Some of the development plans around the track do suggest to me – and, again, I’m not an expert on real estate developments – that the company may have wanted to develop some of the nearby land in a way where that fit well with a card casino / horse track in the same neighborhood. Some of the other planned stuff just sounds like what any developer would want to put in that area though. So, it’s a mix. But, I don’t want to make it sound like Canterbury Park is planning to just sell off these 100+ acres to get some cash. They’re not. They will be left with more people in the immediate vicinity of the card casino and track and with some equity stakes in entities that will provide cash flows for a long time to come. So, while I looked at this as a “sum of the parts” – I don’t expect Canterbury to just take a lump sum of cash for a development that won’t benefit the race track and card casino at all. I expect the track and casino will benefit from the development and Canterbury Park will have an ongoing interest in some of the stuff that’s developed.
Which brings me to capital allocation. This company has increased its share count more than I’d like. Till recently, they had not paid much of a dividend. I think they’re now likely to regularly increase the dividend. But, I could be wrong about that. I see no indication they intend to buy back stock (though they have authorization to do so). Unfortunately, I expect the share count to rise and for this to be somewhere between a minor and major drag on future returns in the stock.
This is basically a family controlled company. The current CEO is the co-founder and son of the chairman. Together they own around 30% of the company. A third co-founder (this company dates back about 25 years) owns another 10%+ chunk. For corporate governance purposes, the third co-founder is treated as independent – though I don’t see it that way. Gabelli is a major shareholder (though not major enough to upset any plans of those 3 insiders). It’s a 5-person board. So, the 3 people who are not the father and son are the ones who make up every committee you’d imagine (compensation, audit, and nominating). The audit isn’t cheap for a company this size. Nor do I have any concerns about the auditor this company uses (they have a local office, they audit other public company issuers, their most recent PCAOB report is clean, etc.). Compared to other gambling companies, closely controlled companies with $50 million market caps, etc. – I really don’t see a lot of problems here. For the most part, the non-founding board members don’t have a ton of ownership in the company. But, I don’t think they’re likely to matter that much in terms of long term capital allocation decisions. The only people I’d expect to have any voice at all are definitely the 2 founding family members, probably the third co-founder, and possibly (as the voice of all the outsiders) Gabelli if they ever want to make a fuss. Otherwise, I find corporate governance and such here pretty typical of a public company. It’s possible I am underestimating the long-term view with this being a family company embarking on a major real estate development that’ll transform the business quite a bit.
My problems here are: 1) They may dilute me more than I want, 2) I know nothing about Minnesota state politics, and 3) I know nothing about Minnesota real estate. Those 3 things could be major factors here. So, while this is a business – the card casino – I like trading at a price I’m fine with, I don’t know if I’d revisit this one or not.
If I was to research this further, my 3 next steps would be:
- Talk to people more knowledgeable about real estate in the area
- Talk to people who follow this stock / Minnesota gaming more than I do
- Talk to management
- Visit the race track, card casino, area to be developed, and competing gambling venues in the area
Those are probably the 4 things I’d want to do before buying this stock.
Geoff’s Initial Interest: 50%
Possible revisit price: $9/share (Down 28%)