Valuing Float in Non Insurance Companies

[Saluki]

Float is a big part of the growth story for Berkshire and Fairfax, but I'm wondering as a thought exercise how you would value it in other companies that have characteristics that resemble insurance float.  For instance, if I buy an airline ticket for my vacation on July 4, I pay now but the flight doesn't happen for a few months, then after the flight happens, they pay for the fuel and salaries etc.  Some companies like Walmart sell product quickly and pay suppliers slowly, so they have free use of that money to build new stores.  Apple and Amazon, last time I checked have similar qualities.  

 

Just thinking out loud here, If it's a fast grower like early Amazon, I guess you can value that float based on the rate of return on their invested capital.  With insurance companies, it's based on what financial instruments they can invest it in.  But with both those options, you have to give the money back. 

 

There's a company that I've posted about before, and I won't mention the name because I'm the only person keeping the post active and I'm not trying to pump the stock, just get some input on valuing something.  If they get a 20 year contract for $20 million, that income is recognized at $1 million a year. But if you get all of the money at the beginning of the contract (assume it's a patent or license or trademark) and you don't have any opportunities to reinvest in growing your business, then what is it worth?  It's not really float because you don't have to pay it back.  And discounting that last $1 million from 20 years to today doesn't make sense either because you don't wait twenty years to get it.  You get it now, and you get to use it and never have to give it back, so from an accounting standpoint it's not income today, but from a practical standpoint it is. 

 

If they don't have any possibility of reinvesting it in the business and you assume that it gets invested in fixed income, then would you value it like a bond?  I assume you would ignore the fact that you recognize $1mm a year as income as the contract progresses, because it would be double counting?  If this is the correct way to value it, then let's a wrench into it.  What happens if you buyback shares? Do you value the effect of the buyback based on the value of the license/patent if you sold it off completely to a third party? Do you value the buyback based on what you paid for the patent/license, i.e. book value?   Do you value it based on the stream of income vs the stock price and whether it's accretive to shareholders?  Or is there some other method that would be better? 

[bizaro86]

30 minutes ago, Saluki said:

Float is a big part of the growth story for Berkshire and Fairfax, but I'm wondering as a thought exercise how you would value it in other companies that have characteristics that resemble insurance float.  For instance, if I buy an airline ticket for my vacation on July 4, I pay now but the flight doesn't happen for a few months, then after the flight happens, they pay for the fuel and salaries etc.  Some companies like Walmart sell product quickly and pay suppliers slowly, so they have free use of that money to build new stores.  Apple and Amazon, last time I checked have similar qualities.  

 

Just thinking out loud here, If it's a fast grower like early Amazon, I guess you can value that float based on the rate of return on their invested capital.  With insurance companies, it's based on what financial instruments they can invest it in.  But with both those options, you have to give the money back. 

 

There's a company that I've posted about before, and I won't mention the name because I'm the only person keeping the post active and I'm not trying to pump the stock, just get some input on valuing something.  If they get a 20 year contract for $20 million, that income is recognized at $1 million a year. But if you get all of the money at the beginning of the contract (assume it's a patent or license or trademark) and you don't have any opportunities to reinvest in growing your business, then what is it worth?  It's not really float because you don't have to pay it back.  And discounting that last $1 million from 20 years to today doesn't make sense either because you don't wait twenty years to get it.  You get it now, and you get to use it and never have to give it back, so from an accounting standpoint it's not income today, but from a practical standpoint it is. 

 

If they don't have any possibility of reinvesting it in the business and you assume that it gets invested in fixed income, then would you value it like a bond?  I assume you would ignore the fact that you recognize $1mm a year as income as the contract progresses, because it would be double counting?  If this is the correct way to value it, then let's a wrench into it.  What happens if you buyback shares? Do you value the effect of the buyback based on the value of the license/patent if you sold it off completely to a third party? Do you value the buyback based on what you paid for the patent/license, i.e. book value?   Do you value it based on the stream of income vs the stock price and whether it's accretive to shareholders?  Or is there some other method that would be better? 

 

Airlines are often not great for float, because their credit card processors often either keep the money, or require large amounts of restricted cash. AirBNB has tons of float along those lines. 

 

For your example, if they have $20 MM of cash from some sort of licensing deal (no ongoing costs) I'd value it at $20 MM. That's exactly like having $20 MM in cash from some other source, except the GAAP shows over many years which doesn't matter, imo. 

 

What they do with it is a capital allocation question, but it isn't different than what would they do with $20MM of cash balances. 

[hasilp89]

I'm not sure I am helping you @Saluki but i'd start by considering the cost of float/leverage and what you earn on it.

additionally I think in most cases the float these non - insurance businesses carry is not long dated. Starbucks card maybe get used in a a few months. Naked wines every 6 months. Expedia 6 months. Pluxee maybe a bit longer or forever if certain amounts aren't spent. So ever if it is free or you are paid to take it, your investment options are limited to ST term Fixed income with lower historical yields.

 

Not sure what business you're talking about (feel free to share :)) but would think similar to @bizaro86 what is your cost to perform the $1m of service/product each year. Assuming you have to deliver $1m of service a year are you able to do that at $1m a year (making the cost 0%) or otherwise.

 

 

[gfp]

You see this sort of "float" with some subscription based businesses where a subscription is sold (cash comes in on day one as total billings) but profit is only recognized over the life of the subscription.  I just analyze these companies based on the cash billings and not on GAAP.  Some companies are required to hold most of that upfront cash on the books (maybe they offer money back guarantees or have to hold some of it as restricted cash or whatever) so it earns 5% today and more or less in other interest rate environments.

 

A well run, growing insurance business isn't really paying back the float - it is just a revolving fund of fungible capital that should grow over time if conditions are right.  Better than the 'un-earned revenue' type of float Anterix would have.

[scorpioncapital]

Float is more useful when inflation is low I'd imagine. Consider the example of the airline. Perhaps they can get 4% for a few months until your flight, but if interest rate is 4%, then inflation is also higher and their costs may be higher 4 months down the road. So it seems to me neutral at best. 

[Saluki]

@hasilp89 if you are curious about it, @gfp guessed it.  Thanks for the input guys.  Yes, the FCC licenses would be worth about twice the market cap if they were used for their original purpose, rather than the current use, for which they had to obtain FCC permission.  Kind of like rezoning a property from commercial to residential.  So I don't think the value to a third party is relevant because you can't use it for the original purpose unless you shut down the business, and now that you started getting contracts, it's kind of hard to undo. If I buy industrial land, pay to have it cleaned up and put up condos, I can't change my mind half way and put a refinery next to it. 

 

They are way behind on their pace, where they thought they would have $1bln in contracts by 2023.  They are about a 1/3 of the way there.  But if that can happen, which is a big if, then if we make some simplifying assumptions like that you are getting 5% on that, that gets you $50mm a year coming in, with no growth and a very small office overhead. As a quick and dirty rule of thumb (I don't remember where I learned it), I think you shouldn't pay a PE more than double the growth rate. So a 10% growth rate should be no more than a 20 PE, roughly speaking, for most businesses. 

 

I don't know what I would pay for something that COULD take in a $1bln and then not grow. I'm sure when the contracts are up, if all goes well you should be able to renew them in 20 years and it should be an inflation adjusted number that mimics the original deal.  So maybe price it like a 5% bond (on whatever the revenue is) with a 20 year duration? But if they do buybacks with the money, then a 10% a year reduction in the share count would be like getting 10% growth, and if you pay 10-20x earnings on that, that would get you to between $500mm and $1bln.  So in terms of enterprise value, it's on the low end of that range of reasonableness, but not a screaming buy or sell. Especially if getting to that range of reasonableness requires some events to occur which are still in the works.  

 

Just thinking out loud here, but thanks again for the input.  

[Spekulatius]

19 hours ago, Saluki said:

Float is a big part of the growth story for Berkshire and Fairfax, but I'm wondering as a thought exercise how you would value it in other companies that have characteristics that resemble insurance float.  For instance, if I buy an airline ticket for my vacation on July 4, I pay now but the flight doesn't happen for a few months, then after the flight happens, they pay for the fuel and salaries etc.  Some companies like Walmart sell product quickly and pay suppliers slowly, so they have free use of that money to build new stores.  Apple and Amazon, last time I checked have similar qualities.  

 

Just thinking out loud here, If it's a fast grower like early Amazon, I guess you can value that float based on the rate of return on their invested capital.  With insurance companies, it's based on what financial instruments they can invest it in.  But with both those options, you have to give the money back. 

 

There's a company that I've posted about before, and I won't mention the name because I'm the only person keeping the post active and I'm not trying to pump the stock, just get some input on valuing something.  If they get a 20 year contract for $20 million, that income is recognized at $1 million a year. But if you get all of the money at the beginning of the contract (assume it's a patent or license or trademark) and you don't have any opportunities to reinvest in growing your business, then what is it worth?  It's not really float because you don't have to pay it back.  And discounting that last $1 million from 20 years to today doesn't make sense either because you don't wait twenty years to get it.  You get it now, and you get to use it and never have to give it back, so from an accounting standpoint it's not income today, but from a practical standpoint it is. 

 

If they don't have any possibility of reinvesting it in the business and you assume that it gets invested in fixed income, then would you value it like a bond?  I assume you would ignore the fact that you recognize $1mm a year as income as the contract progresses, because it would be double counting?  If this is the correct way to value it, then let's a wrench into it.  What happens if you buyback shares? Do you value the effect of the buyback based on the value of the license/patent if you sold it off completely to a third party? Do you value the buyback based on what you paid for the patent/license, i.e. book value?   Do you value it based on the stream of income vs the stock price and whether it's accretive to shareholders?  Or is there some other method that would be better? 

For prepaid services you just look at FCF instead of earnings and you are basically done.

 

As far as insurance float is concerned, it’s accounted for in the ROE imo. However, you need to look at float separately, if the newer if s power of the float will change dramatically as a result for a change in interests rates or if management decided to invest the float differently.

 

The change in interest rates has been a huge tailwind for is insurance companies because they get far higher returns on fixed income, but also have MTM on their existing fixed income positions in many cases.

[dwy000]

There are different kinds of float that would have different values.  Insurance companies float is a reserve for "what if" whereas a subscription business needs that float to cover the cogs over the life of the subscription.  Then there's subscriptions without a defined product delivery schedule - like Amazon Prime.  You can also consider companies like Costco and Dell to have massive float in that they have negative working capital.

 

Id just value FCF but be very careful because for those with negative working capital or subscriptions, if growth turns flat or negative for a while you could have a really large outflow required (well beyond the revenue decline) and they need to be able to fund it.

[Saluki]

Good points!  Yes, I was probably overcomplicating it.  FCF makes the most sense.  Unlike a subscription, there is no pro-rated refund for cancelling, and you get the money up front, so if you minus the overhead (they had 80+ people in the head office, last time I checked), there should be no depreciation expense because it's more like a brand than a factory that wears out. 

 

In retrospect, the setup on ATEX looks better now than when I first initiated a position 3.5 years ago and posted about it. I notice a LOT of great idea/great technology small caps and most of them don't go anywhere. There is a YouTube channel called PlanetMicroCap, which has a lot of these story stocks. They are interesting, but 90% of them don't make any money. If I knew someone who needed to get rich fast, I would tell them to find a confident straight white guy with a full head of hair, buy him a new suit and have him go on there as your CEO and you will get a lot of money with an idea and a pitch deck.   If you wait until they are cash flow positive, that cuts out about 80% of them with a single filter.  If I used that filter, I would have come in later to Coupang.  I think they turned positive earnings at around $16 instead of $27 when I first started buying it.  I'm very impressed the tech at Enovix, which I first noticed at $14, and is now below $8, so I'm glad that I only bought a tiny amount of shares to keep an eye on it instead of building a position. So although ATEX is about where I started buying it, if I had waited until more cash was coming in than going out, that cash could easily have been a double in the SP500 and it would be a better process which cuts out a few winners, but also gets rid of every loser. Maybe this is another decision that I should add to the Atul Gawande investment checklist? 

[ValueArb]

One thing to look for when you are reading new 10Ks is a deferred revenue liability in the balance sheet, esp. a long term deferred revenue liability. Its almost always some form of float. 

[gfp]

I happened upon this article and it seemed relevant to @Saluki 's thread.  It's a few years old.  The author of the post also authored a book on the early years of Berkshire Hathaway fwiw.  He seems to collect vintage financial reports.

 

https://mcdinvestments.substack.com/p/another-industry-on-which-berkshire

[ValueArb]

Jason also has an excellent podcast where he reviews annual reports from historical companies that is surprisingly engrossing. He's gone over how Ford, GM, and NCR started and its extremely interesting to see how innovations that you never even thought about (the cash register) created huge value for the economy.

 

https://podcasts.apple.com/us/podcast/the-10-k-podcast/id1690744153

[Saluki]

23 hours ago, gfp said:

I happened upon this article and it seemed relevant to @Saluki 's thread.  It's a few years old.  The author of the post also authored a book on the early years of Berkshire Hathaway fwiw.  He seems to collect vintage financial reports.

 

https://mcdinvestments.substack.com/p/another-industry-on-which-berkshire

 

Thanks @gfp, this was an interesting read.  I remember looking at Chicago Bridge and Iron, another EPC company a few years ago and I was not comfortable with the fact that deciding whether a project is 30% or 40% done involves a lot of subjective decisions and thus they can decide how much to report and it would be too easy to manipulate earnings. The float like aspects of the upfront payments and what you get to do with the money didn't even occur to me. 

 

Sometimes looking at things a different way will give you a better appreciation for the value of something.  When Amazon bought Whole Foods, some people wondered why they bought a struggling high end retailer.  I remember Prof G (Scott Galloway), who predicted it, saying that it one move Amazon bought refrigerated distribution centers in every major US city. 

 

@ValueArb I have Jason's book sitting on my boookshelf and I haven't cracked it open, but I will.