Curious to see what others on the forum think about this, as it’s something that perhaps will be a more frequent conversation within equity valuation circles (and maybe even sooner for those in EMs).
Here’s a rather extreme case: say that starting tomorrow Amazon starts accepting Bitcoin as a form of payment on their site, and within a few months their sales settled in Bitcoin account for 20% of revenues. Presumably, the riskiness of their business would rise, in that the volatility of their revenues when translated to USD would increase markedly.
How would you go about accounting for this added layer of risk when building, say, a DCF model for Amazon? Would you try and capture it in the discount rate used for the cash flows, or maybe explore some sort of scenario analysis that models for crisis in cryptocurrency markets and all the uncertainties regarding regulation, cyber security, etc?
Financial literature is pretty good at dealing with this when you’re talking about valuing companies whose revenues come from different currencies issued by government bodies… but not so much with something decentralized like what cryptos are predicated on.
I don’t see why the source of the volatility of the currency matters. Seems no different than any other volatile currency to me. But I’m a far cry from a math wiz
only really matters if they are holding bitcoin on their balance sheet. in the case of overstock they are. don’t know if you would create scenarios for valuation though. today’s price is the most accurate price. i suppose you could create some scenarios where BTC goes to $20,000 and where BTC goes to $0 but that is just to gauge the effect on the stock in those scenarios. you wouldn’t use scenarios to find a fair value for OSTK, just to gauge risks.
BitCoin revenues should be given special treatment, not just because it is a cryptocurrency, but because it has such high volatility. BitCoin prices dropped 10% on Monday, and that is just a normal move for this currency. If GBP or JPY were to move 10% regularly, this would create havoc on valuations for companies which rely on international trade. There are few currencies with anywhere near the volatility of BitCoin, and this might justify treating BitCoin as a special type of asset.
Furthermore, we should account for settlement lag. Merchant payments don’t happen instantaneously. Sometimes, they need days or weeks to deliver payment. As far as I know, there is not forward market for BitCoin. So, this creates unhedgeable risk from potential BitCoin price fluctuations.
We might also want to consider certain regulatory tail risks. If, for instance, BitCoin became a non-viable payment method overnight, due to regulations or other issues, would the 20% market share of customers who rely on BitCoin also disappear?
All this suggests that a higher discount factor should be applied to BitCoin revenues to reflect the higher risk of this payment method.
Yea, your post summarizes to me all the reasons why the discount rate would need to reflect the increased idiosyncratic risks that come along with accepting BTC as a form of payment. What I’m curious about is how to go about this exercise of scaling up the discount rate to adequately reflect these risks.
That would be the case if payments were immediate and seamless. However, when you deal with thousands of vendor invoices and delayed payments with no forward market, you will have real settlement risk. Plus, we can’t ignore the risk of business disruption from a highly unstable exchange medium. When CHF depreciated abruptly in 2015 due to the Swiss central bank unexpectedly changing their cap relative to EUR, regional trade was disrupted and Swiss firms recorded huge losses. Imagine if currency fluctuations like this happened not once a decade, but once a week.
So, where BitCoin is concerned, would a currency with near 100% annualized volatility create additional risks if it was a large portion of a company’s payments? Of course. Would this necessitate model adjustments to take into account this additional risk? Only an incredibly reckless analyst would say no.
I’m not a research analyst, so I don’t know what is market convention for this sort of discount. However, CFA teaches us that cost of equity is estimated as {risk free rate + Beta*Market Risk Premium}. So, if we know the risk free rate, MRP, and use historical volatility to estimate a “beta”, then that might be an ok estimate. It will look pretty high though.
And that’s where conventions like the CAPM would fail. That cost of equity obviously won’t do much good if your estimated historical beta are from the days that cryptocurrenices were not a part of your company’s revenues. I have no clue what the “rigorous” work around to that might be though. Anyways, I don’t think there is a convention as of yet for dealing with companies whose revenues are either partially settled in BTC or in the more extreme case held on the balance sheet like Overstock.
No, I mean estimate a risk adjusted discount rate for BitCoin revenues, using only BitCoin historical volatility. All other cash flows in the model can be unmodified (to the extent that BitCoin doesn’t cause risk to those additional revenues).
but Bitcoin payments to most retailers are paid in BTC and received in USD. my understanding is the only risk takers are the customer who initially owns BTC and Bitpay who acts as a medium? the retailer only ever holds dollars. i think only OSTK holds BTC so only it would potentially have a BTC volatility risk. that said, as BTC can only go down 100% and it can go up 1000%, is this additional volatility offset by extremely bullish BTC scenarios? maybe. i wouldn’t adjust OSTK’s valuation for BTC risk as it is a two way risk that is nearly impossible to compute and requires huge qualitative assumptions. i do agree that if a large portion of your customers are BTC users, a BTC crash could effect sales volumes but a BTC rally could boost sales volumes as well so pricing again requires major qualitative assumptions.
You should discount volatile cash flows at a higher discount rate… even if there is symmetric upside. This is the basis of capital asset pricing model. You lose more utility by being down 50% than you gain from being up 50%. Risk neutral pricing has important financial implications (for instance, in Black-Scholes style derivatives pricing), but certainly not in this sort of asset valuation.
I am thinking about the forward price risk of BitCoin payments. If prices are denominated in USD, but just settled though BitCoin (I don’t know if this is the case - maybe it is), then what I mentioned about forward risk specifically should not apply. However, there will still be a risk of merchants not being able to settle payments in a timely manner, due to BitCoin price fluctuations, as well as customer demand volatility as their BitCoin wealth fluctuates in value.