Alico (ALCO): A Florida Orange Grower Selling Land, Paying Down Debt, and Focusing on its Core Business
Alico (ALCO) is a landowner in Florida. The company is – or is quickly becoming – basically just an owner of citrus groves that produce oranges for use in Tropicana orange juice. The majority of the land Alico owns is still ranch land. The company has about 100,000 acres in Florida. Of this about 55,000 acres are ranch land and 45,000 acres are orange groves. The book value consists almost entirely of the actual capitalized cost of the orange trees on the land. The land itself – with a few exceptions caused by recent purchases – is held at unrealistic values on the balance sheet. For example, the company has sold ranch land at more than $2,500 per acre that was carried on the books at less than $150 an acre. So, the situation here is similar to two other stocks I’ve written up in the past: Keweenaw Land Association (KEWL) and Maui Land & Pineapple (MLP).
There is one write-up of the stock over at Value Investors Club. You can go over to VIC and read that write-up. It gives background on the history of the company that Alico itself doesn’t really talk about in either its 10-Ks or its investor presentation. The company has recently tried to get its story out to investors. There is now an investor presentation. There have also been a couple quarters of earnings calls.
The investor presentation has a slide that includes the company’s estimate of the fair value of the land it owns versus the enterprise value. On this basis, the stock looks cheap. However, it doesn’t look incredibly cheap. And I’m somewhat unsure whether a value investor should look at the stock as just a matter of enterprise value versus likely market value of the land. But, I’ll start there, because other write-ups of the stock will almost certainly be focused around that investor presentation slide that lays out the company’s enterprise value versus the likely fair market value of the land.
ALICO owns 55,000 acres of ranch land. (For the purposes of this write-up, I’m using some out of date numbers not updated for land sales and cash receipts – however, they basically would just net out: less land, more cash / less debt). The company puts an estimate for fair value of that land at $2,000 to $3,000 an acre. Ranch land I’ve known of in other places goes for similar amounts to that. About $1,000 to $3,000 an acre. The company has sold plenty of ranch land recently. And much of it has been sold in the $2,000 to $3,000 range. So, that implies a pre-tax value of $110 million to $165 million for the ranch land. However, almost all of any land sales not put into a “like kind” asset to defer taxation will end up taxed at very, very high amounts because nearly 100% of the sale will be a capital gain. Also, some of this land seems to me to be encumbered with debt. Alico isn’t totally explicit about what land is encumbered with debt and what land isn’t. The company does give the total amount of debt that has land as collateral backing it. It gives this number in both acres and in dollar amount (of the loan). However, it does not break out what lending is backed by citrus groves and what lending is backed by ranch land. So, it’s possible – this is just me guessing – that in some cases, Alico may not be allowed to use proceeds from certain land sales for anything but debt repayment and possibly the purchase of additional land (that the lender agrees to).
My interpretation of the borrowing / asset situation here is a little different. I think, basically, the ranch land and the debt are “non-core” and they basically cancel out. Alico sees itself as an orange grower and is moving in that direction. It is selling off ranch land at the same time it is buying orange groves. It also did a deal in the last year to start managing an orange grove (7,000 acres – I believe these are “net” acres that actually produce oranges, there are probably support acres not counted in this) on behalf of another landowner in Florida.
Why do I say the ranch land and the borrowing offset? One, they are probably pretty close in terms of net debt versus value of the ranch land. Both could be around $150 million. Two, it is not difficult to borrow long-term against orange groves. The company is already borrowing long-term (it’s about 2/3rds fixed rate borrowing and 1/3rd variable rate borrowing) from Met and Prudential. Based on other “permanent crop” agricultural loans I’ve seen and what this company has historically borrowed at – I think the fully leveraged “cost of capital” for the orange groves would potentially be long-term fixed rate borrowing at 3-4% after-tax (so, a bit higher pre-tax) equal to about one-third of the fair market value of the groves. I think you’ve definitely got additional borrowing capacity whenever your total debt / fair market value of orange groves is under 30%. The company says the groves are worth $8,000 to $10,000 an acre. So, it’s probably possible to borrow like $2,500 to $3,000 per orange grove acre at like 3-4% after-tax on a continuing, long-term (and effectively permanent) basis. One warning: when you re-borrow, you’ll be paying whatever long-term rates of interest are at that moment.
I mention this because it’s unclear the company intends to operate with like no leverage if it sells off the ranch land. I think a lot of people looking at the situation would say the company is selling off ranch land and paying down debt. This is true. But, I don’t think it’s a liquidation. We can see some sold land is encumbered (that’s mentioned at times). And the company sometimes has “restricted cash” that seems to be tied to the land sales. I think a realistic long-term view of the company’s future progression would be to assume ranch land is sold off over time and the proceeds are put into citrus grove purchases where possible and then to debt repayment to the extent no tax-deferrable “like kind” purchase can soak up the capital freed up in the land sale. This could take a very long time. But, for a long-term investor, the state the company will probably get to is one in which it has sold off the ranch land and paid off the debt.
Having said that, I don’t assume that an orange grove operator would use no leverage. But, I also don’t think it’s safe and cheap to operate with more leverage than about 1/3rd of the value of the groves. After that, you’re just like any other corporate borrower. At low loan-to-vale ratios, you have a lot of collateral to borrow against even if you’re an otherwise weak credit. So, over time, it may be that not all the debt is ever paid off. However, the debt will eventually be repaid and re-borrowed in a way that shifts the ranch land off the books and shifts whatever debt remains to being tied to the orange groves.
So, I don’t think a liquidation analysis shows you what you’re really buying here.
Another point to mention is the dividend. Alico has paid a dividend for close to 50 years now. It started upping that dividend a bit in recent years. And then it upped it a lot recently. Alico isn’t doing special dividends, buybacks (except for one really big tender it did a couple years ago), etc. So, the decision on the regular dividend rate is probably tied to the cash flow from the orange groves. The ranch land produces essentially no income. My best estimate based on looking at segment reporting is that ranch land produced $1 million to $2 million in EBIT at times. However, the company discloses one aggregates lease, one oil lease (it owns the mineral rights under its land for about 90% of its acreage) and some grazing rights leasing. I don’t think the grazing rights leasing contributed any real income. The company says the mining royalties aren’t “material”. But, something did produce some cash flow which might be immaterial at the corporate level but explain the small production of income from ranch and other non-citrus land we’ve seen at times. As best I can tell, whatever recurring income was coming in from the land subsidiary (as opposed to the citrus subsidiary) has been due to some other leases – probably “mineral rights” which may be aggregate mining or something else like oil, etc. There had been plans for a water project. However, the company abandoned those and sold the land it would’ve needed to complete that project. So, forget the water stuff you’ll read about in the Value Investors Club write-up of Alico. That’s dead.
As a result: I think we can basically remove whatever EBIT we’ve seen from the ranch land (again, it’s minimal) and remove the debt and remove the ranch land.
This leaves just an orange growing operation on the “value” side of the equation and just the market cap (forget the debt, so EV doesn’t matter) on the “price side”. Market cap is $230 million. There’s 45,000 acres of orange groves. So, $230 million / 45,000 acres = $5,100 an acre. This is less than the company says the land is worth ($8,000 – $10,000 an acre). The market cap / fair market value of the citrus land is therefore theoretically 50-65%. There’s Ben Graham’s famous “one third margin of safety”. And that’s why value investors might get interested in this as an “asset play”.
But, let’s talk about it as an actual business.
Customer concentration is extreme. However, I don’t think it’s relevant. The company sells 90% of its output to Tropicana under 5-year type contracts. These contracts aren’t very meaningful in my analysis, because I’m assuming Tropicana is the most logical customer for Alico and Alico is one of the most logical suppliers for Tropicana. These are commodities. And the contracts are not very fixed. The contracts probably just codify the likely microeconomic situation anyway. The way these contracts work is that a market price (at the time the contract is signed) for oranges sold per pound of useable material (pounds solid soluble) establishes a price range that is fixed. Outside of this price range, Tropicana and Alico are basically sharing 50% of the rise or fall in the market price of oranges per pound.
You can find current and historical pricing, production, estimates for next season, etc. at the USDA. You can calculate Alico’s cost structure for yourself using cost of goods sold and pounds sold from the company’s 10-K. I’m not going to go over those numbers here. Instead, I’ll focus on something you might miss under GAAP accounting.
What’s more important is the way Alico’s cost structure differs from the contract structure. The variable cost of producing oranges for orange juice is low. My best guess is that no more than 30% of the cost of Alico’s product is actually tied to factors like production levels. About 70% of the “cost” Alico shows in its per pound numbers, cost of goods sold, etc. is really a fixed cost of operating the groves. The 10-K has a line that is super explicit about this point. The groves cost what the groves cost. It is only costs like hauling that vary with production levels. So, if no oranges were sold in any one year (which will never happen, but bear with me), the company would still show an expense equal to about 70% of its cost structure. So, the variability of pricing and production levels of oranges is greater than the variability built into the contract. In other words, Alico’s earnings are cyclical and are tied to the amount of revenue it can generate at market prices for its oranges each year. The revenue depends on quantity of output (which varies a lot depending on weather conditions during the year) and the market price of oranges. The market price of the oranges depends on inventory levels for oranges, exports of oranges from Brazil to the U.S., exports of oranges from Mexico to the U.S., and the level of oranges produced in the U.S. (basically in Florida, though California and Texas probably produce some oranges too).
Due to COVID, not from concentrate (the type of orange juice Tropicana produces and Alico is an input for) demand rose and orange juice demand continues to be at an elevated level. COVID didn’t disrupt the orange harvest. But, production declined a bit for weather reasons. Obviously, production can’t be increased for oranges the way it can be for commodities planted each year. An orange tree has to be planted about 4 years in advance of producing any fruit. Production then peaks around 8 years after planting. The tree continues to produce. Based on the way Alico depreciates its groves, it seems like groves have up to a 25-year producing life (and almost 5-year non-producing life). As a result, increases in demand and decreases due to weather move the price for the oranges. Demand moves instantly. But, there can’t be any immediate adjustment in supply.
The one thing I would worry about here more than contracts with Tropicana is tariffs. The U.S. has tariffs on orange juice imports from other countries. So, the competitive position of oranges grown in Florida versus oranges grown in Brazil does depend in part on political decisions inside the U.S., trade negotiations, etc. This is not a totally free market.
Alico has done a couple things recently that might increase earnings of the groves. It cut operating expenses a few years ago. It also started planting more than a normal amount of trees. So, the cap-ex you’re seeing is not all maintenance cap-ex – some of it is growth cap-ex. This started a couple years ago and is going to continue for another year or so. This cap-ex – the GAAP element to this is capitalized and then depreciated over many years – will only drive additional production 4 years after the planting. And the impact of the additional plantings will peak about 8 years after the original investment.
With any real estate backed stock like this you want to ask how much “owner earnings” really are absent capital appreciation. Sure, if there’s a lot of inflation and the groves are worth $8,000 an acre – at 10% inflation, the land value is producing $800 an acre in capital gains. You’ll do fine. But, what if there isn’t much inflation in land values. Or, what if you have to compare an asset play like Alcio to a good business you could buy instead.
Then, the question becomes…
Are you really earning a decent cash “rent” while you own the land?
What is the cash owner earnings value of the crop itself?
This is hard to tell. EBITDA is not low. It’s been around $25 million at times. It could be higher once recent plantings mature. However, the “depreciation” part of this expense is somewhat real. The groves – this includes roads and water infrastructure and other stuff needed to support the actual productive acreage – has an original cost on the books of $300 million. Even if you depreciate $300 million at a 30 year lifespan, that’s $10 million of annual depreciation. And, with some inflation, the cost of replacing groves is actually higher in future years than the original (nominal) cost you’re depreciating. So, EBITDA here is not like EBITDA on an apartment or even – I’d say – on timberland. It’s more real than that.
You could try to measure free cash flow. But, that’s very cyclical. Both pricing and production move by 20% or more in a given year. And cap-ex is very discretionary. You can under-plant or over-plant for a few years and not see the full impact of that for 4-8 years. You could trust management. You could look at plantings in one year versus another. But, really, if you’re not an expert on this – and I’m not – it’s going to be hard to calculate normal, average, “full cycle” annual free cash flow.
There are some hints about the productive value of groves in cash terms. If you look at the deal Alico signed just to manage orange grove acreage for a competitor – the management fee charged is like $170 per acre per year. That’s my estimate. Alico just says it is reimbursed for all costs and then gets a management fee. But, we can see EBIT from this operation is $300,000 per quarter and they are managing 7,000 acres – so, $300,000 / 7,000 acres = $43/quarter which is the same as $172 a year.
If you’re willing to pay someone $170/year to manage your apartment building, your hotel, your timberland, your farmland, etc. – then, the underlying asset is probably capable of producing quite a bit more than that. This suggests – and the warning here is AT CURRENT INTEREST RATE LEVELS – that a value of $8,000 to $10,000 per acre may be consistent with the actual underlying cash flows over a full cycle. However, it’s certainly not clearly cheap. I mean a management fee – basically guaranteed profit – of $170 on something valued at $8,000 or more is just 2% of the asset’s value. Is a 2% of asset value management fee reasonable – yeah, it might be. The land might be worth $8,000 – $10,000 and also be capable of producing a decent yield at that market price. For example, if the owner is making a 60/40 split with the manager here – that’s a 6% annual pre-tax yield, an 80/20 split is even better, and so on. I mean, the management fee does seem somewhat in line with the market values Alico is claiming.
However, “somewhat in line to me” doesn’t really afford any precision sufficient to promise we can tell the difference between the land being worth $10,000 or $5,000. And that’s your entire margin of safety in the stock right there. Also, I do need to make the warning that Alico has bought some orange acres at less than $8,000. Some acres have gone for closer to $5,000. I don’t know if those were smaller and less efficient acres, if they were less dense groves, etc. But, it’s not like we’ve never seen evidence of a purchase price below $8,000 per acre for orange groves in the area – we have.
So: is the stock cheap?
It doesn’t look especially cheap now. It may be cheap versus earnings in a good earnings year. We might see a good earnings year next year. It is cheap versus the company provided fair market value of the land. But, I don’t think this thing is liquidating. So, this isn’t like buying a dollar for 65 cents.
Is the business good?
Depends. I think the business is getting better here. The company – there’s a whole past history with a group of investors, lawsuits, a dissolution of a partnership that controlled a lot of the stock, changes in the board, etc. I’m not getting into – has a strategy it’s executing on that seems much smarter than what Alico had been doing historically.
Is the asset good?
In times of inflation, yes. In times of no inflation, maybe not. I don’t know if the orange business here can match returns in more asset light and better situated “franchise” type operating companies. Things with moats. Things with good returns on capital. Actual free cash flow producers.
But, is it a better investment than a lot of real estate?
That’s possible. This could be a better way to own real estate than a lot of publicly traded vehicles. There’s a real cash generative business on top of the land. The company is narrowing its operations and focusing on operational efficiency in one product in one small part of the country. It is managing as well as owning. It’s a real operator.
Alico doesn’t look super cheap. But, for a publicly traded land play, it’s not expensive relative to your other options. And capital allocation is moving in the right direction.
The company is doing a big investor relations type push. The management – you can listen to the calls – is really talking up the company, giving guidance, etc.
The stock is kind of under the radar now. It might not be in the future.
At present, I have no real view one way or the other on management. I have looked into the history of the people involved in this one. But, that’s not a topic I feel I know enough about to talk intelligently on.
Alico might be a stock to keep an eye on. Like I said, it’s very cyclical. And investors might get overly excited or depressed in years where both pricing and production happen to move in the same direction.