1/ Just finished reading @jlppfeffer's paper "An (Institutional) Investor’s Take on Cryptoassets" (Link: s3.eu-west-2.amazonaws.com/john-pfeffer/A…) Here's my summary.
2/ The first section is on utility tokens
3/ The value of cryptoasset utility tokens will at equilibrium not decouple from M=PQ/V where
4/ PQ is the total cost of the computing resources required to maintain the blockchains.
5/ V is the velocity at which tokens move and M is the monetary value of the network (i.e. it's market capitalization)
6/ This overall value of most utility cryptoassets will be low because:
7/ 1. The ability to fork will prevent economic rents and moats - "The ability to fork protocols is a great way to maximise utility to users but a bad way to create and sustain economic rent for token holders"
8/ 2. The Fact that transactions don’t require humans will drive high velocity, driving down value M in terms of M=PQ/V.
9/ This means that “the disruption of traditional networked businesses by decentralised protocol challengers will represent an enormous transfer of utility to users and an enormous destruction of market value”
10/ So basically he is arguing that if you accept M=PQ/V and look at reasonable estimates for PQ and V then M is likely to be much lower than current valuations would imply for most utility cryptoassets (e.g. ETH)
11/ The second section is on cryptocurrency as money. The two functions of money discussed are a store of value and means of payment.
12/ Historically there is not an overlap between something being used as a store of value and means of payment. No one stores value in their Paypal or Visa and no one uses their physical gold as a means of payment. (Cash is the one exception that does both)
13/ In cryptocurrency, it’s likely the same will be the case and interoperability will mean one coin can function as a store of value while another will function as a means of payment.
14/ "I lean towards the view that specialisation combined with protocol interoperability is a more likely equilibrium than one coin to rule them all.”
15/ This could look something like Bitcoin as a store of value (b/c of it’s decentralization and fixed supply schedule which make it a more secure store of value but inefficient for Payments)
16/ with Ripple for payments (it’s more centralized and has an uncertain supply schedule but has cheaper transactions and better bank support).
17/ In this view, a utility crypto asset (like ETH) may be seen as working capital to be minimized rather than a store of value.
18/ What’s the potential equilibrium outcome? 
19/ A deeper dive into payments...
20/ A crypto payment network will be worth much less than it’s centralized counterparts b/c at mature equilibrium, the value of such a token would be M = PQ/V...
21/ where PQ is the computing cost of running the chain (something like Visa’s annual IT budget times some coefficient to adjust for the inefficiency of a decentralized architecture) and V is some very high number. Low PQ/High V means a very low M.
22/ This means crypto payment networks may be great for users, but probably will produce poor returns for investors.
23/ a deeper dive into the Monetary Store of Value use case...
24/ It’s likely that one cryptocurrency will become the dominant non-sovereign monetary store of value.
25/ BTC is currently the best positioned to occupy that role.
26/ A good estimate of that cryptocurrency’s value would incorporate the gold market and reserves held in fiat.
27/ "If Bitcoin were to become the dominant monetary store of value cryptoasset, based on my total mature network value estimate of USD 4.7 – 14.6 trillion, it would be worth approximately USD 260,000 – 800,000 per BTC fully-diluted at maturity.”
28/ An investment with 26-80x upside requires a probability of success between 1% and 4% to be a positive net expected value investment though it’s unclear how long this would take.
29/ That does not mean it could not all go to zero. 
