Editing Coinbase

We are launching coverage of Coinbase Global (COIN) with a Buy rating and a $306 price target for 36% upside. We believe COIN brings 1) leverage to an ecosystem that has seen strong growth driven by increasing adoption of digital currencies, 2) a leading consumer platform with strong customer acquisition trends as well as a robust and rapidly growing institutional business, 3) an attractive business model that thrives on elevated cryptocurrency volatility, and 4) significant opportunities to add additional features and capabilities. While we believe that the continued success or failure of cryptocurrencies as an asset class will inevitably determine COIN’s longer-term fate, we believe COIN represents a blue-chip way through which to invest in the development of the ecosystem, powered by 1) its careful approach to regulatory compliance and KYC/AML, 2) its crypto-native technology stack and deep talent pool, and 3) its role as an innovation hub for new crypto endeavors as a result of its status as critical industry infrastructure, and its venture arm. While we believe the core business today offers an attractive growth profile with the potential to drive high levels of profitability, we see significant white space for new
initiatives to drive more stable and recurring revenue streams to complement the core trading business over the longer term.

Key Takeaways:

# The best way to gain exposure to the expansion of the crypto-native ecosystem: COIN’s goal is to be the primary financial account for consumers to access the crypto-native ecosystem. While the crypto-native ecosystem today is still nascent, we are watching 1) the adoption of stablecoin-based payments, 2) innovations in DeFi, or decentralized autonomous blockchain applications, and 3) the adoption of non-fungible tokens (NFTs) and the creation of markets for tokenized, real-world assets, as potential drivers of crypto-native ecosystem expansion in the near term. If meaningful parts of the economy can transition to blockchain and crypto-native technology over time, we see significant opportunity for COIN to benefit from its status as a critical element of the financial infrastructure for the ecosystem.
# User growth to drive strong growth in transaction revenues over time: COIN has grown its users at a 35% CAGR over the past 3 years, reaching 56mn in 1Q21. With ~90% of revenues coming from retail trading, in the near term we see continued strong growth in users driving solid organic growth for the business. While we expect growth to continue, over time we expect 1) retail commission pricing pressure (where we highlight COIN’s current pricing and a handful of potential catalysts for compression), and 2) crypto market conditions to heavily influence the trajectory of revenues over time. On this latter point, we believe investors are too focused on the impact of prices and less focused on the impact of volatility, which we believe is the bigger driver in the near term. We leverage historical regression analysis to build a factor model for COIN revenues to estimate what COIN’s three guidance scenarios imply for the top line in 2021 (we believe $4.6bn for the “low” scenario and $9.5bn for the “high” scenario). Lastly, we sensitize our 2023 transaction revenue forecasts for various price and volatility environments.
# Significant white space for additional products: While just 4% of COIN’s revenue now comes from non-trading activities, we believe subscription and services revenue has the potential to see outsized growth relative to the core run rate of the business as COIN rolls out additional ancillary services over time. We believe staking revenues are poised for particularly strong growth given 1) the recent addition of Cardano (ADA) to the platform (which roughly doubles the current TAM on COIN’s platform for staking revenues), and 2) the transition of Ethereum to proof-of-stake (ETH 2.0). We also see collateralized lending (similar to margin lending) as a significant opportunity, as we project a ~$75bn lending opportunity for the crypto industry. We also expect strong growth in custody revenues driven by strong growth in institutional AUM. Lastly, over a longer period of time, we see entering the NFT space as well as blockchain-based payments as additional areas of expansion.
# Expect a somewhat volatile margin trajectory: COIN’s cost base is largely fixed, with headcount-related expenses supporting its technology, customer services, and back office driving the majority of the expense base and directly variable transaction expenses comprising less than 15% of revenues. COIN management has said that over the course of various crypto pricing cycles they will scale their business to operate at roughly breakeven over time. That said, our model does not take an explicit view of crypto prices and holds crypto market cap stable over time, and we expect to take a mark to market approach as prices fluctuate. Given COIN’s high levels of profitability at current crypto prices, we model ~30-40% EBITDA margins over time. That said, on our current expense base, we see breakeven levels of profitability if prices were to fall roughly 50% from today’s level, although continued reinvestment in the business will likely put an upward bias on the breakeven levels over the longer term.
# Valuation: We value COIN using a EV/Sales approach given the high growth rates and volatile margin profile of the company. We look at a range of comps, including market structure-oriented companies such as MKTX, TW, CME, and ICE, high-growth consumer-facing online brokers, such as IBKR, as well as payment-oriented companies such as SQ and PYPL. Our 12 month price target is based on 13x EV/sales on 2023E net revenues, which is roughly in line with more market structure-oriented exchanges and implies a market cap of $81bn and a 12 month price target of $306, for 36% upside over the next 12 months. Over time, increasing acceptance of crypto as a means of exchange, if that were to occur, would imply higher valuation multiples more in line with higher-growth payment companies, in our view. Key downside risks include 1) lower crypto pricing / volatility, 2) lower commission rates, and 3) crypto regulation.

==Best way to gain exposure to the expansion of the crypto-native ecosystem==

Coinbase aims to be the primary financial account for users to access the
“cryptoeconomy” or what we refer to as a crypto-native ecosystem, and we believe
Coinbase is the way to gain exposure to the development of the crypto-native
ecosystem given 1) its status as the largest fiat on-ramp in the industry for both retail
and institutional clients, serving as the starting point for many users to interact with
crypto, and 2) its focus on safety and security and regulatory compliance, providing one
of the largest and most secure, institutional-grade custody solutions for crypto assets in
the industry. That being said, we emphasize that our base case forecasts largely assume
the status quo in terms of the ecosystem, and this section is largely geared towards
highlighting why Coinbase is well positioned should the crypto-native ecosystem
continue to develop. As we show in Exhibit 2, Coinbase has a strong track record of
consistent market share gains in the amount of crypto assets held on its platform. While
other fiat on-ramps exist, we believe COIN’s position as the number 1 on-ramp from the
US financial system makes it a compelling play on the development of the crypto
ecosystem. Coinbase’s 56mn users represent a sizable user base to expand the product
over time, and we believe that if activity in the crypto-native ecosystem takes off,
Coinbase will increasingly be able to capitalize on its leading market position. In the
following section we highlight Coinbase’s unique position as the top fiat on-ramp to the
crypto ecosystem, its strong position as the top institutional custodian (which should
continue to support strong growth in the institutional business), as well as a number of
potential catalysts for further development in the crypto-native ecosystem and providing
some ways to help size the opportunity.

That being said, we stress that the crypto-native ecosystem is still fairly undeveloped.
We would describe the crypto-native ecosystem as all economic activity that is
conducted using crypto-native technology that is separated from the fiat banking rails.
Today we believe the bulk of the activity in the ecosystem is happening at the
cryptoexchanges, which help facilitate transactions in crypto currency and charge a fee
for these services. As a result, much of the ecosystem is market structure-oriented in
nature as the various participants in the industry, such as Coinbase, have helped build
the financial infrastructure on which the ecosystem can develop. Looking forward,
however, we believe 1) the adoption of crypto-denominated payments, particularly
through stablecoins, 2) further proliferation of non-fungible tokens (NFTs), and 3)
considerable white space for the development of “DeFi” applications, or decentralized
applications that run via smart contracts on blockchains such as Ethereum have the
potential to drive considerable expansion of the economic activity in the crypto
ecosystem. Should this occur, though this is not part of our forecast, we believe that the
ecosystem has the potential over time to drive meaningful amounts of activity and
commerce. 

'''Retail: Well positioned as number 1 on-ramp to the Crypto ecosystem'''
Coinbase offers three primary retail platforms for buying and selling Cryptocurrencies:

* Coinbase: Coinbase is a simplified, user-friendly interface that allows the buying and selling of cryptocurrencies geared towards more “everyday” use. While the Coinbase platform has a simpler user interface than Coinbase Pro, the fees are higher, sometimes as high as 3-4% of the transaction value. Coinbase also offers several other products to customers including a Visa debit card that is funded by cryptocurrency funds held in the account. COIN serves as the custodian for the cryptocurrencies held on the platform, meaning the company centrally manages the storage of private keys.
* Coinbase Pro: Coinbase Pro provides users more advanced access to COIN’s cryptocurrency exchange, where the company manages a centralized order book and allows for various order types and trading strategies. The platform also offers more information on market liquidity and depth. The Coinbase Pro is geared towards institutions and more advanced retail traders but is available for anyone to use. Fees are much lower on the Pro platform, starting at 50bps and decreasing to 0-5bps in an exchange style maker-taker pricing plan with volume-based tiers.
* Coinbase Wallet: Coinbase wallet allows users to store and send/receive cryptocurrency to/from other users. The Coinbase wallet app is a user-managed wallet and the user’s private key, the cryptocurrency equivalent to an account password, is stored on the device itself. If the private key is lost, the users funds can never be recovered although the company has introduced cloud-based backup solutions to address this risk. Coinbase offers this solution in part to allow Coinbase users to interact with other parts of the crypto ecosystem such as decentralized applications.

'''Competitive dynamics are heating up with increasing competition from fiat native firms'''

We expect strong user growth to continue over time on the retail side as Coinbase leverages the accessibility of its platform to the mass market. COIN has grown its verified user base over time at a 35% CAGR since the beginning of 2018 with monthly transacting users increasing to 6.1mn in 1Q21 and averaging 5% of total verified users over time. As shown in Exhibit 8, COIN appears to dominate the retail mind share for full-service crypto exchanges, measured by mobile app downloads, with a distribution platform more in line with other mass market fintech offerings such as Square’s Cash App, Paypal, and E-Toro. We believe this demonstrates the competitive position relative to other crypto native firms, making them well positioned to continue to acquire new customers who wish to interact with the crypto ecosystem. That said, a number of investors have highlighted the increasing competition from fintechs, as well as from traditional brokerages (Fidelity is a large digital asset custodian and SCHW and IBKR are both reportedly working on crypto offerings). While these firms do not have the technological advantage of being fully crypto-native companies and often rely on outsourced solutions for crypto, we believe their significant investment resources derived from their non-crypto-related activities makes them a competitive threat over the longer term, although this would likely require significant investments or acquisitions to catch up to COIN’s considerable lead.

'''Institutional: Top-tier custody and execution services are set to drive increasing institutional market share'''

On the Institutional side, we believe Coinbase is one of a handful of venues to offer full
service institutional capital markets services for the crypto ecosystem and believe COIN
has a significant moat against the competition as a result of its top-tier custody offering
as well as the 2020 acquisition of Tagomi, which powers the institutional platform and
drives a compelling multi-venue trade execution offering. This has allowed institutional
assets to grow to comprise over 50% of the assets on the platform over a relatively
short period of time. Given the large sums of money at stake and the risk of theft or loss
of funds due to the inherent nature of cryptocurrencies as digital bearer instruments, we
believe custody is a central part of the conversation with institutional clients. Coinbase’s
widely lauded custody platform, with a track record of having never lost customer funds,
is a strong competitive advantage relative to other less proven platforms, in our view.
We believe increasing institutional adoption in the latter half of 2020 and into 2021 has
been a material driver of the acceleration in the institutional business and would
highlight the significant flows into the Grayscale investment trust (GBTC), a publicly
traded investment vehicle that has reached $38.4bn in AUM (Exhibit 10), and which custodies the underlying bitcoin at Coinbase. That said, we see increasing interest
among other institutions, particularly from traditional financial institutions (through white
label offerings), fintechs, wealth management platforms, family offices and endowments
as catalysts to drive further growth in the institutional business over time. While the
revenue generation on institutional assets is fairly low given the lower trading take rates
and negotiated rates on custody, we see significant runway for growth and view the
institutional part of the business as important for the normalization and adoption of
crypto assets.

'''Charting the development of the crypto-native ecosystem: Moving from store-of-value to crypto as a means of exchange'''

While Coinbase today largely provides crypto trading services, the company has a goal
of being users’ primary financial account for the “cryptoeconomy,” or the crypto-native
ecosystem as we refer to it, leading many to ask what exactly is the state of the
ecosystem and where are we in its development. While our base case assumes largely
the status quo in terms of the state of the ecosystem, to answer this question, we find
it helpful to look at cryptocurrencies and the activity they facilitate on a spectrum
between “store-of-value” activities and “means-of-exchange” activities. In the classical
definition of money, currencies tend to fill the role of being both a store of value and a
medium of exchange in the economy (we would also include “unit of account” for
completeness). The crypto ecosystem today is heavily skewed towards “store-of-value”
types of activities. This largely includes the pricing, trading, and custody of crypto
assets. However, since inception, blockchain technology has allowed for peer-to-peer
payments between wallets, which form the basic building block for more
means-of-exchange type of activity in the crypto ecosystem. So far, however, we have
yet to see significant amounts of non-capital markets-driven payment activity or
widespread adoption of crypto payments. This is due to a handful of factors, including 1)
the high volatility of crypto prices, 2) capacity constraints given the relatively low
transaction throughput of most blockchains relative to traditional payment rails, and 3)
the challenges of driving network effects for adoption from the ground up. While there is
still much progress to be made to reach widespread acceptance, and we note that
broad acceptance is not our base case, below we highlight several potential catalysts
over time to ease barriers to adoption. We believe each of these catalysts provides an
opportunity for more value-added means of exchange activities in the ecosystem. 

'''1) The rise of Stablecoins: Blockchain-based, fiat denominated, instantaneous, cross-border payments'''

Stablecoins are digital tokenized representations of fiat currencies that maintain a peg to
an underlying currency (e.g., 1 stablecoin = 1 USD) by holding reserves on a one-for-one
basis in an audited bank account. The supply of these currencies is variable and they can
be minted/burned based on demand by either redeeming a stablecoin for the underlying
deposit or paying dollars/fiat in exchange for receiving a stablecoin on the blockchain.
Unlike Bitcoin and other cryptocurrencies, the value of stablecoins does not fluctuate,
providing a constant value to use as a means of exchange. While stablecoins’ values are
tied to fiat currency, they also have the convenient “programmable money” attributes of
cryptocurrencies, allowing for blockchain-based payments and the integration of fiat
currency into blockchain-based smart contract applications.
We believe that the price stability of stablecoins likely makes them strong contenders
for near-term adoption of cryptopayments, and makes them more of a pure play on the
benefits of blockchain technology as a whole without the associated price volatility of
cryptocurrencies. Tether (USDT) and USD coin (USDC) are the two largest stablecoins,
accounting for the majority of the market (Exhibit 13). The outstanding supply of
stablecoins has grown rapidly over the last year (Exhibit 12). That being said, we believe
much of the activity in stablecoins today still facilitates more store-of-value activities, such as serving as a reserve currency for market markers on various exchanges who
wish to keep idle capital denominated in crypto, as well as representing stand-ins for fiat
currencies on unregulated trading venues (i.e., users on Binance transact in BTC/USDT
currency pairs instead of BTC/USD). Recent events, however, have pointed to continued
progress in the adoption of stablecoin denominated payments, such as a recent
announcement by Visa that the company had begun processing payments denominated
in USDC from the cryptoexchange Crypto.com, allowing the crypto-native firm to pay
third parties directly in cryptocurrency without the need to convert crypto into fiat
currency to make payments. Stablecoins provide a low friction payment mechanism
with fast settlement times, allowing users to bypass fiat payment rails (ACH, Fedwire,
SWIFT, etc.) that are not available 24/7 for instant settlement. Stablecoins can also run
on multiple blockchains, making them portable to more scalable blockchains in the
future that are better suited to high-throughput payment situations.

On this last point, we note that people often raise the issue of transaction throughput as
a gating factor for the adoption of cryptopayments, which is something we are less
focused on. While it is true that most of the mainstream blockchains such as the Bitcoin
blockchain have relatively low throughput compared to a fiat payment networks, there
are numerous projects (examples: Lighting Network, ETH 2.0) aimed at reducing traffic
congestion on blockchains as well as alternative blockchains under development aimed
specifically at high volume situations. In other words, the industry is working on
upgrading from “dial-up” to “broadband” over time, to use an internet analogy.

'''2) The rise of DeFi: decentralized smart-contract enabled money applications can drive real activity in the crypto-native ecosystem'''

The term DeFi (short for decentralized finance) in a general sense pertains to all aspects
of decentralized cryptocurrencies in the sense that blockchain technology facilitates an
inherently decentralized financial system. That said, over the past 2-3 years, DeFi has
taken on a more specific meaning pertaining to cryptocurrency applications that are built
on top of smart contract platforms, which allow developers to write code that executes
on the blockchain and determines how transactions are executed. This has given rise to
several new platforms that allow various activities to take place in a fully decentralized
manner where no centralized authority governs the application’s operation. Most DeFi applications today (ranked by “Total value locked”, similar to Coinbase’s assets on
platform metric) fall into two main categories: 1) decentralized exchanges (DEX),
performing functions similar to Coinbase for buying and selling crypto currencies
(examples: Uniswap, Sushiswap, Curve Finance), and 2) lending protocols, which allow
for decentralized borrowing and lending of cryptocurrency at programatically determined
interest rates (examples: Maker, Compound, Aave). However, there are various other
DeFi platforms facilitating other activities such as derivatives and payments.
The development of DeFi applications is still in its infancy, and relative to the nearly
~$2+ trillion market cap for crypto currencies, the total value of crypto on DeFi
applications is still relatively low at just ~$67bn. In addition, the current use cases
involving exchanges and collateralized lending (the latter of which we would argue more
closely resembles margin loans than true lending) don’t do much to move the needle of
crypto-native activity more towards the means-of-exchange direction, in our view. That
being said, we believe they represent important proofs of concept for more complex
applications in the future. One of the most important features of many of these
protocols, in our view, is that many DeFi applications have native “governance tokens”
(for example, UNI for Uniswap, COMP for Compound) that allow the holders of the
token to vote on changes to the protocol’s operations. In addition, it is possible to
configure these protocols to pay the holders of the governance tokens a share of the
revenue generated by the protocols (for instance, the token holders for a decentralized
exchange might receive a share of the trading fees or a share of the interest on a
decentralized lending protocol). These two components of governance and a claim on
the income stream provide a framework for thinking about decentralized autonomous
organizations (DAO), the crypto version of a corporation, and their associated
governance tokens, a crypto version of an equity security. While there are a number of
regulatory hurdles to overcome in order to normalize this, we believe this emerging
framework represents one of the larger longer-term opportunities for the development
of the crypto-native ecosystem.

'''3) The rise of non-fungible tokens: decentralized tokenized ownership of goods'''
Much of the focus around NFTs centers on digital artwork and collectibles and some of
the recent sales prices for NFTs, such as the $69mn sale price at Christie’s for an NFT representing a piece of art by the artist Beeple, or the significant volumes of NFTs being
sold on the NBA Top Shot platform. NFTs at their core leverage blockchains to tokenize
the ownership of unique goods and reduce transaction costs by simplifying the
exchange of ownership. In our view, it is the crypto analogue to the title to a vehicle or a
house: It’s not the physical asset itself but it does verify the holder of the token as the
owner of that asset. Beyond the current use case of digital art, we believe NFTs have
the potential to facilitate high friction and low liquidity transactions, such as in real
estate, physical artwork, and certain areas of the credit market, such as leveraged loans,
to list a handful of examples. While it is still early days, and there remain a number of
legal hurdles to overcome, including how NFTs are regulated and how to handle the
fractionalization of NFTs over time (which could lead to more security-like tokens), we
believe NFTs represent another potential avenue of development for the crypto-native
ecosystem over time. 

'''A bottom-up approach to the crypto TAM'''

We believe drawing direct parallels to the fiat world as a model for the market
opportunity for crypto is challenging, as we believe crypto has the potential to alter the
economics, reduce spreads, but also expand the market size of financial activities in the
traditional financial industry. That said, a number of investors have pointed to the various
profit pools of traditional financial services activities to contextualize the activities that
crypto could facilitate in the future. We separate financial activities into more
store-of-value oriented activities and more means-of-exchange oriented and lay out the
rough market size of each to give a sense of the relative magnitude of each. In general,
while the store-of-value TAMs are much larger than the exchange-oriented activities, we
believe the revenue margin / take rates tend to be much lower on store of value
activities than in medium of exchange-type activities and pricing / take rates on store of
value activities have tended to be more deflationary than in payments.

'''Forecasting assets-on-platform'''

On the next page, we show the mechanics of our model for forecasting
assets-on-platform over time. We break up growth in assets-on-platform into 1) organic
growth, defined as a net new asset growth rate, representing new deposits of assets
onto the platform over time, and 2) market impacts, which are driven by changes in the
value of the crypto assets held on the platform. While Coinbase does not provide this
breakout of the flows on the platform, we attempt to back into the number historically
by estimating the impact of market levels on assets by looking at the change in the
overall crypto market capitalization. We attribute the remaining piece to organic growth.

* Retail NNA growth: Retail net new assets (NNA) growth looks to have slowed somewhat in recent quarters. COIN has said that they expect retail to be the slower growing platform in the near term given this segment is more mature than its institutional business, but has not spoken to recent growth trends. We highlight a few potential explanations including large increases in asset levels, leading to profit taking and withdrawals from the platform or transfers to self-custody solutions, but given the lack of detailed quarterly trends, the organic growth rate of retail assets is something we are watching in the near term.
* Institutional NNA: Our math implies that institutional NNA growth has been very robust recently, with significant new deposits of assets on the platform. We think this makes sense the last 6 months have seen a number of headlines around institutional investors taking stakes in crypto, leading to chunky deposits of assets on the platform. In line with COIN commentary, we believe the organic growth rate in institutional will remain very robust, leading to mix shift in assets-on-platform over time.
* Market impacts and implied market share: We note that our baseline forecast doesn’t take an explicit view on crypto prices, and mechanically we are modeling a flat crypto market capitalization over time. That being said, we are also modeling significant amounts of new assets-on-platform over time, with assets increasing to nearly half a trillion dollars by 2025, which under a flat market scenario, would imply market share approaching 25% of all crypto assets. The value of a currency is derived in part from the public’s perception of that currency as a viable store of value and means of exchange. Thus, it is difficult to imagine a world in which COIN’s user growth remains at ~20%+ over time off its current 56mn verified user base and public acceptance of crypto doesn’t increase, and therefore, we feel that the flat market assumption in our model could overstate the market share gains.

==Strong user growth should drive strong top-line growth==

Switching gears from the macro to the micro as it relates to Coinbase: while we expect
strong growth in assets and customers on both the retail and institutional side of the
business and view COIN as one of the best plays on the development of the
crypto-native ecosystem, given the low commission rates on institutional volumes and
the early stage of development for the crypto-native ecosystem, we believe retail
trading will continue to dominate the P&L and make up the majority of COIN revenue
for the foreseeable future. In this section, we provide our thoughts on retail pricing
pressure, where we believe that pricing for retail crypto trading will likely follow similar
trends to the brokerage industry over time, where commissions compressed over
several decades as other revenue streams developed, and expect the existence already
of free trading offerings in the industry will be hard to ignore as competition for retail
relationships heats up. Additionally, we provide insight into some of the market share
dynamics using the reported volumes of various crypto exchanges, where we show that
COIN has 20-30% market share for trading volumes among other “fiat exchanges”
(which comprise 30% of reported volumes) and roughly 7% of overall volumes. Finally,
we demonstrate why we believe that while the prices of crypto are obviously
important for revenue, in the near term the volatility of crypto asset prices is a
very important, if not more important driver of revenue, as customer activity
levels is highly correlated with crypto asset volatility over time. We use our price
and volatility framework to flesh out COIN’s guidance for 2021 and provide a
revenue sensitivity under various market conditions.

'''Competitive landscape for crypto trading: we see a fragmented volumes landscape with a large share of volumes at unregulated exchanges'''

Data sources on the crypto ecosystem remain somewhat fragmented. That being said,
we have leveraged data from theblockcrypto.com, a crypto research firm and data
provider to analyze the landscape of trading volumes. The data can be accessed here.
The volume data is broken down into fiat exchanges, or exchanges that allow deposits
of fiat currency from the traditional banking system and allow crypto/fiat trading pairs and crypto-only exchanges, or exchanges that do not accept USD deposits and largely offer crypto/crypto currency pairs. Among the fiat exchanges, Coinbase compares
favorably with the highest market share for volumes of any player, with its next closest
competitors, Upbit and Kraken roughly half the size of Coinbase. That said, when we
include the international exchanges Binance, Huobi, and OKex, Coinbase emerges as
the 4th largest exchange, with Binance appearing to be by far the largest, accounting for
roughly 43% of the total reported volumes relative to Coinbases’s 7%, and likely
representing COIN’s most significant competitor outside of the United States.

By definition, the lower Know-Your-Customer (KYC) requirements and wider range of
unregulated activities allowed on these platforms have the potential to drive more
activity on those platforms than onto Coinbase, which performs identify verification on
all users and prioritizes strict compliance with US regulations before adding additional
features. We view Coinbase’s strict compliance policies as a competitive advantage, as
the rapid increase in the crypto market is likely to attract more attention from
governments, in our view, which has the potential to cause problems for unregulated
exchanges with lax KYC policies and looser requirements for listing new
cryptocurrencies (Bloomberg reported in March that Binance was under investigation by
the CFTC for allowing US residents to trade unregulated crypto derivatives). The lack of
access to the US banking system and US customers for these companies is also a key
competitive advantage for Coinbase. In addition, this approach helps the company
establish friendly relations with rulemaking bodies, which is particularly important given
Coinbase will likely play a significant role in lobbying for more tailored regulation for
cryptocurrency in the future. Lastly, it makes the United States somewhat of a captive
market and acts as a barrier to international, non-compliant exchanges. That being said,
Coinbase’s strict compliance approach has meant that less-regulated exchanges such as
Binance have generated strong retail brand recognition outside of the United States and
likely represents a significant competitor as Coinbase expands internationally. 

'''Retail pricing likely to be under structural pressure over time'''

Coinbase makes roughly 50bps of revenue on the notional value of volumes traded on
its platform, and on retail volumes it receives roughly 140bps on average, relative to the
something on the order of 1bp or less that the online retail brokers receive in payment
for order flow for their free trading offerings in the United States. This has led a number
of investors to ask about the potential for pricing pressure over time. Coinbase
management has said that similar to the pricing trends we have seen in traditional
financial services, over a longer period of time it is likely they will see pressure on
prices, but that there is nothing in the near term they see driving significant
compression and that they do not seek to compete on price.

We expect that pricing will compress over time, but similar to the pressure we saw in
the brokerage space over time, the pressure is likely to be episodic, unpredictable and
interspersed with times of relative stability. Below we highlight a handful of catalysts for
price compression over time. Lastly, on the following page, we have attempted to
compare pricing across various crypto venues. Note that given the variability of pricing
across exchanges, as well as add-on fees, and the complexity of some pricing plans
(particularly COIN’s base retail pricing) we would say the following chart is something of
an oversimplification in some cases, but is good for illustrative purposes.

* Free trading offerings: A number of firms have free crypto trading offerings today that have the potential to put competitive pressure on commission charging exchanges. In the fiat world, free online trading is generally supported by payment for order flow, which is effectively charged to retail investors through the bid-ask spread, but is generally extremely low (order of 1bps for retail equity volumes). In crypto, there is not the same market structure for payment for order flow, but there still exists the ability for companies to set bid-ask spreads for customers wider than they can otherwise source the inventory and thus earn a small fee. Thus “free” trading offerings have emerged where the trading cost is embedded in the bid-ask spreads of the crypto currencies. Further proliferation of crypto trading services marketed as free to the retail investor is likely to put incremental pressure on
cryptoexchanges such as COIN to lower prices over time.

* Platform harmonization: This is largely a Coinbase-specific risk. Coinbase offers two platforms, Coinbase and Coinbase Pro, to retail customers. While there are some difference in terms of the capabilities of each platform, in general, Coinbase Pro provides significantly lower prices for the customer, starting at 50bps, than the Coinbase platform, which has fees of 50bps up to 450bps and for small dollar transactions can have fees as high as 10%. Coinbase management has talked to the potential to see better integration of the multiple products into a more unified pricing tier, and has specifically mentioned reducing pricing on smaller dollar transactions, which are more likely to incur higher fees in the current pricing structure. We believe any restructuring of the various retail offerings is likely to result in lower pricing. 
* Reduced barriers to entry: One reason prices are high in the industry is the relatively young age of the crypto industry and the barriers to entry posed from the new forms of technology that crypto requires. We also highlight the difficulty of interacting with the crypto-native ecosystem at scale as a result of the difficulty in securing banking relationships to make cryptocurrencies easily convertible into fiat currency. For instance, Coinbase is at a premium price point to nearly every other major crypto exchange in the United States but is also one of the easiest cryptocurrency exchanges on which to open and fund an account via ACH as a result of Coinbase’s focus on regulatory compliance and their strong banking relationships. We believe this makes the pricing differential somewhat sustainable in the near term, but over time, reduced barriers to entry and higher competition from other firms is likely to result in pricing pressure. 

'''Delving into the 2021 revenue guidance'''

On its 1Q21 earnings call, COIN provided a handful of scenarios that it used to illustrate 
how it plans for its volatile and difficult-to-predict revenue model. While its expense
guidance was relatively prescriptive, the top-line commentary left some room for
interpretation, as COIN provided thoughts only on where monthly users could be in
different crypto pricing and volatility environments. Coinbase’s business is highly
impacted by crypto prices, and many investors have asked what these scenarios imply
for revenue and thus the bottom line in 2021. As shown in Exhibit 30 and Exhibit 31, the
correlation of Coinbase volumes and revenue with the market capitalization of crypto is
fairly high. That said, we do not believe the crypto market cap is the only driver of
profitability and below, we seek to tease out the relationship between revenues and
volatility and make the case that the sensitivity to prices is perhaps less than it appears,
with volatility being a very important driver. We then use the historical relationship to
derive ARPU assumptions to apply to COIN’s 2021 guidance.

'''Crypto asset volatility is a large swing factor for revenue'''
COIN talks about their revenue trends in terms of “monthly transaction revenue per
monthly transacting user” (we will abbreviate as ARPU). We also note we are calculating
this metric based on total transaction revenue, not just retail. In Exhibit 32, we show the
historical trends in ARPU over time, which appears to be highly sensitive to volatility
over time. Note that COIN didn’t provide quarterly income statements for 2018 and so
we are using indicative take rates on their disclosed volume and Monthly Transacting
User (MTU) numbers to estimate ARPU in 2018 and provide more historical context. As
shown in the exhibit, ARPU has been highly correlated with the volatility of crypto prices
over time. Note that the impact of volatility is particularly powerful, as not only do users
tend to trade more, but more users tend to trade in general, demonstrated by the
correlation between MTUs and volatility in the past (Exhibit 33).

'''How to think about the impact of higher crypto market caps on volumes'''
While volatility is a significant contributor to quarterly results, we believe there is a
general perception that higher crypto prices in general directly impact revenue trends.
While we do believe there is a relationship, we do not believe the sensitivity is as linear
as investors perceive given the relatively low correlation of ARPU and crypto market
cap. Instead, in Exhibit 34, we show that the correlation is actually closer to a
logarithmic relationship, implying decreasing sensitivity of revenue as prices increase
and higher sensitivity as prices fall.

Based on the relationship laid out above, in the following exhibit we leverage COIN’s
2021 guidance for MTUs and the various volatility and price scenarios to show where
the top line could emerge. After layering in our assumptions for subscription and
services and applying COIN’s relatively prescriptive guidance for the line items of
expenses, we derive a range for 2021E of $4.6-9.5bn in revenues and $1.7-$5.0bn in
EBITDA (we emphasize that these company-based scenarios are separate from our own
forecasts for 2021, which are in the column headed “GS” in the following exhibit). 

'''How do we think about out year forecast sensitivity?'''
Lastly, in the following discussion we take the earlier derived regression on historical
volatility and crypto market cap to derive a range of outcomes for the top line under
different scenarios. Because both the number of users and the amount traded per user
are linearly correlated with volatility and just logarithmically correlated with price, we
find much greater sensitivity of revenues to volatility than with market cap over time.
Exhibit 37 shows what our model predicts for transaction revenues under various
scenarios and demonstrates the very wide range of outcomes that exists, although we
note that the volatility environment appears to play a greater role in revenues than the
absolute level of crypto market cap. Exhibit 38 back-tests this model historically (using
the same estimated values for 2018 and 1Q21 that we have used throughout this
section) and finds a fairly high correlation with reported revenues.

==Significant white space for additional products over time==

While we expect transaction revenue to be the largest driver of the P&L for the
foreseeable future, we believe the subscription and services line will be the primary
place in which the development of the crypto-native ecosystem manifests itself
overtime. As of 2020, these line items were just 4% of overall revenue, with custody
and staking representing roughly 2/3s of the overall pie. These line items are generally
more correlated to the asset levels as opposed to the flows, or the trading volumes, on
the platform, and because the volumes / turnover of assets is a multiple of the assets
on platform and occur at higher weighted average take rates, we expect trading revenue
to continue to comprise the bulk of the revenues for the foreseeable future. That said,
should the crypto-native ecosystem build out and the profile of business activities
transition from capital markets / speculative / store-of-value activities to more
commercial / economic / means-of-exchange activity, we would expect to see greater
emphasis on the subscription and services fee growth. Below we review each item and
then show how we are deriving our forecast.

'''Custody: Likely to grow with institutional assets over time'''

Coinbase is the world’s largest digital asset custodian for cryptocurrencies and is the
administrator of several high profile institutional accounts, including the custodian for
the Grayscale Investment Trust, the largest publicly traded Bitcoin investment vehicle
with $38.4bn in AUM as of March 2021. Coinbase also sourced Tesla’s BTC purchases in
1Q21, and has facilitated the purchases of BTC by Microstrategy. In addition, Coinbase
has referenced university endowments as Coinbase institutional clients, at a time when
several large Ivy League university endowments have been reportedly building direct
BTC positions.

Custody for crypto assets has some of the same record keeping requirements as
traditional securities custody in the financial sector but the real crux of crypto custody is
the secure management of crypto wallet private keys, which serve as digital passwords
for crypto wallets. Cryptocurrencies are bearer instruments, and whoever holds the
private keys is the owner of the digital assets in the wallet, and thus the storage of
private keys serves as both a cybersecurity function as well as a custody function.
Coinbase was the first digital asset custodian to receive SOC 1 and SOC 2 reports,
which are industry standard reports certifying the quality of internal controls.

We view COIN’s custody offering as a cornerstone of the business, as COIN has
invested significant resources in de-risking the storage of crypto and becoming a
regulated and fully compliant destination for institutional clients to interact with the
crypto-native ecosystem. We believe the average institutional client’s first real step to
engage in the holding of crypto begins with custody, and view COIN’s status as the
largest custodian as a key competitive advantage for winning institutional business, as
well as a key driver of investors’ faith in the Coinbase platform, particularly if we see
high profile custody failures in the industry in the future.

'''Staking: increased adoption of staking and ETH 2.0 transition to drive staking revenue growth in excess of asset growth'''

The second largest component of Coinbase’s subscription and services revenue is
staking revenue. Staking is conceptually similar to mining, and the revenue is derived
from the rewards that accrue to the validators (analogous to miners) when they validate
a new block on the blockchain. Staking does not happen with all cryptocurrencies; it is
dependent on the “consensus mechanism” that the various blockchains use to add new
blocks to the blockchain. Bitcoin uses a “proof-of-work” consensus mechanism,
whereby miners compete on computing power to solve a mathematical puzzle, and the
first to solve the puzzle gets to add the next block to the blockchain and is rewarded a
preset amount of newly minted coins (each new BTC block mined currently produces
BTC 6.25 for the miner). (On an unrelated note, we would note as a “fun fact” that the
first transaction on a new bitcoin block, representing the payment (BTC 6.25 currently)
to the miner in exchange for adding the next block to the blockchain, is called the
“coinbase transaction.”) In contrast, blockchains that run on proof-of-stake allow
validators (instead of miners) to “stake” their holdings of a currency to the blockchain.
The blockchain protocol then allocates each new block pseudo-randomly (the more
tokens one stakes, the higher probability of being selected to validate the next block) to
the next validator. The validator receives the transaction fees associated with that block.
This yields an interest-like income stream for validators who stake their cryptocurrency
to a proof-of-stake blockchain, with the amount of income proportionate to the amount
of currency staked, and the interest rate defined by the supply and demand of
transaction fees and validators on the network.

Coinbase hosts a number of proof-of-stake currencies on their platform, including
Algorand (ALGO), Cosmos (ATOM) and Tezos (XTZ). In addition, in 1Q21, COIN added
support on the retail platform for Cardano (ADA), which currently is by far the largest
proof-of-stake currency (Exhibit 45), and we would expect staking to follow at some
point in the future, as staking is already supported for the other PoS currencies above
and Coinbase already announced in 4Q20 they would be adding staking support for ADA
on the custody platform soon. Coinbase earns staking revenue by taking a 25% cut of
the staking rewards from its users in exchange for providing the infrastructure for users
to stake their PoS holdings, as well as covering the potential for losses of user funds if
Coinbase incorrectly validates a new block (this is the punishment mechanism for PoS
blockchains that ensure validators do not incorrectly validate transactions).

As an aside, we are interested in seeing where this 25% cut of the revenues goes over
the longer term. On one hand, it is a seemingly very large take-rate and one might think
it has the potential to compress over time, particularly when the fee is high compared to
competitors such as Kraken and Binance, which don’t charge staking fees at all. On the
other hand, this activity seems eerily similar to bank-like intermediation by taking idle
customer funds and putting them to productive economic work (i.e., if you have an 8%
yield and you pay customer 6%, it resembles a 2% net interest margin). In the
traditional banking system, the best banks are those that share the least amount of total
interest income with customers through lower deposit costs, which is usually because
customers are going to them for more value added services rather than being lured in
by higher deposit rates. These companies tend to capture the greatest percentage of
the revenue generated from deploying those funds. Thus, as the economic utility of the
Coinbase platform to its users increases over time, we could see an argument for its
share of these sorts of yield generating activities increasing, although this is by no
means our base case and would probably be accompanied by lower yields. 

'''Transition to ETH 2.0 could be significant driver of staking revenue going forward'''

As a final point on staking revenues, Ethereum, the second-largest cryptocurrency,
currently runs on a proof-of-work consensus mechanism, but given some of the scaling
problems and the proliferation of smart contract and DeFi applications running on the
Ethereum blockchain, there is a transition underway to migrate Ethereum to a
proof-of-stake mechanism (dubbed ETH 2.0), which is less energy-intensive and
potentially more scalable. While it remains uncertain how successful this transition will
be, COIN is already operating an ETH 2.0 node and has a wait-list for ETH staking. Given
that ETH is currently 14% of the assets on platform, this has the potential to provide a
significant uplift in staking revenues should this occur, which is generally expected by
the sector to happen over the next year or two. Exhibit 46 provides an estimate of this
benefit based on our Assets on platform forecasts, which shows that depending on the
percentage of customer ETH that is staked, ETH 2.0 staking could generate
~$120-210mn in annual revenues over time at current market levels.

'''Earn: DTC marketing tool to drive increased awareness likely to grow as the crypto-native ecosystem expands'''

“Earn” is the smallest crypto-native income stream in the Subscription and services
line. Earn revenues are generated when users participate in educational programs to
learn about new tokens and activities in the crypto ecosystem. These are often
sponsored by third parties and users generally receive rewards based on the completion
of the educational campaign, often in the form of the token they are learning about, with
Coinbase receiving a commission for sponsoring the activity. We view this as a form of
direct-to-consumer marketing for crypto to increase the awareness of newer parts of
the ecosystem.

'''Interest income: largely driven by customer cash balances and interest rate environment.'''

Coinbase generates interest income off of customer cash balances in a very similar way
to how Charles Schwab generates net interest income off of customer cash. COIN’s 1Q
earnings release showed that roughly 3% of assets on platform at the end of the period
were in the form of fiat currencies, which are held in traditional bank accounts with
COIN’s bank relationships. These balances earn interest income, which, similar to an
online broker monetizing customer float, COIN keeps as revenue. Given the zero
interest rate environment, the amount that Coinbase earns on these balance is fairly
low. That said, in the following exhibits we show how this revenue stream would
increase if rates were to rise, resulting in higher interest rates on customer cash. 

'''Lending opportunities: We see significant opportunities to provide collateralized lending that could be a double-digit uplift in EBITDA over time'''

Increasingly, centralized participants in the crypto ecosystem have been providing
crypto-collateralized lending to holders of cryptocurrency. Our sense is that currently,
the majority of this lending activity most resembles either capital markets-oriented
portfolio leverage like margin lending, or short financing. In our view the addressable
market for this type of activity is fairly limited (at least relative to the size of the crypto
market cap), as it largely resembles over-collateralized asset transformation rather than
true economic borrowing activity. If we look at margin financing and short interest
(which is correlated to securities lending activity) in the traditional financial system,
these items account for a relatively low percentage of the outstanding market cap of
stocks (~2% margin loans relative to NYSE and NASDAQ market cap in the United
States, roughly 2-2.5% average short interest in the S&P 500). Thus, we believe the
addressable market for crypto collateralized loans is likely in the mid single digit
percentages of the outstanding crypto market cap. Using this range for the addressable
market for crypto-backed lending, Exhibit 50 shows the impact that building out lending
products on the platform could have over time on our base case EBITDA estimate, for
which, depending on the degree of penetration and the amount COIN holds on balance
sheet (we believe the opportunity may be too capital intensive for COIN to fund the
entirety of these loans), we see the potential for a ~10% uplift in EBITDA over time from
the build-out of lending products.

Transitioning to more means-of-exchange style lending: That all being said, we
believe the much larger opportunity over the long term is true credit creation in the
crypto-native ecosystem, where crypto is lent in an uncollateralized fashion, or is
secured by hard assets (potentially tokenized versions on a smart contract platform?) or
working capital. In addition, the significant price volatility of crypto assets makes
crypto-denominated debts challenging, and if we do see this type of activity arise, it is
more likely to occur in stablecoins, which are pegged to fiat currencies rather than
floating crypto. True lending is inherently more risky, as the lack of liquid collateral
creates material amounts of credit risk and subjects the lender to a potential loss of
funds, and we are not sure that COIN has a desire to be in the lending business. That
being said, the extension of credit is a core part of the economy and when done
correctly, has the potential to facilitate true commerce in the space and intermediate
borrowers and savers, an activity we think has the potential to emerge in the space over
time. We note in Exhibit 49 that the US banking system is >60% loan/deposit ratio,
which in our view is indicative of the longer-term lending potential for a crypto
ecosystem.

'''Other opportunities'''
Payments likely multiple enhancing opportunity over time: Crypto payments has
yet to see mass adoption as a result of technological barriers, price volatility, long
transaction confirmation times, and low transaction throughput of the highest profile
blockchain networks. We showed in Exhibit 16 that means of exchange activities tend to
have both large TAMs while also benefiting from higher/more stable take rates over
time, with less volatility than store of value activities, which can be buffeted by a
number of cyclical factors. We believe payments is the most multiple-enhancing activity
COIN could pursue, as the valuations for payment companies tend to be much higher
than that of capital markets businesses. COIN is currently experimenting with a Visa
debit card that links to a Coinbase account funded with cryptocurrencies. While this
particular activity is largely immaterial today, we believe it represents progress in the
means of exchange direction and demonstrates progress on the acceptance front by normalizing using crypto for transactions.

Non-fungible Tokens (NFTs): Digitized and fractionalized ownership: The recent
proliferation of NFTs or non-fungible tokens demonstrates the ease with which unique,
scarce assets can be created and transferred on the blockchain. While much of the NFT
space revolves around digital artwork, we believe NFTs have the potential to expand to
include digital representations of physical objects such as real estate and artwork as
well as intangible assets such intellectual property. Given the crypto infrastructure that
NFTs are built upon, we would expect Coinbase to expand support for NFT custody over
time, with the potential to provide access or support for NFT exchanges as well as the
issuance of NFTs. This also seems to us like a logical space for a bolt-on acquisition.

==Fixed cost base likely to result in volatile margin profile==

More than 90% of COIN’s revenues come from transaction revenues with very limited
variable costs (transaction costs average low double digit percentages of transaction
revenue). Thus, COIN’s margin profile is characterized by relatively high fixed costs,
largely derived from technology and headcount related expenses. This factor, when
combined with a volatile revenue stream, is likely to lead to a high degree of earnings
volatility over time. Coinbase management operates the company under an assumption
of continued volatility of crypto prices with large positive and negative pricing trends
over time, resulting in highly variable margins. Because of this dynamic, the company
has messaged that they expect to run the company at breakeven over time through a
pricing cycle, with episodes of relatively higher pricing resulting in elevated profitability
(such as now) and episodes of lower pricing (“crypto winters” per the company)
potentially driving operating losses. That said, our model does not take an explicit view
on crypto prices and assumes flat crypto prices over time, resulting in much more
steady profitability than is likely to occur. We expect that the market will likely take a
mark-to-market approach with crypto prices, which is likely to result in significant
estimate volatility over time. 

* Tech and development costs: Tech and development are associated with costs related to engineering and product development and the operating costs of keeping the platform running. We believe the majority of these costs are fixed in nature but expect some variability given the platform-driven costs, which likely have some sensitivity to activity on the platform.
* General and administrative: G&A costs include general support costs and other non-operating expenses such as back office, legal, HR, executive comp and other miscellaneous costs laid out in the company’s S-1. We consider these costs largely fixed in nature, although there is likely some variability in incentive compensation related to firm performance.
* Sales and marketing: Marketing costs includes customer acquisition expenses and advertising costs. Historically the company has not spent a large amount on sales and marketing (just 8% of expenses in 2020) but going forward, the company has guided to a significant increase in marketing costs to accelerate customer acquisition and stimulate its verified user base. We believe this ramp is ongoing and believe the exit run rate of sales and marketing into 2022 will likely exceed the full-year guidance of 12-15% given elevated revenues in 1Q21 and back-loaded builds in expenses. Note this assumes a moderation of revenues from 1Q consistent with the middle scenario of guidance for the top line.

Transaction costs are likely to vary with transaction volumes: Coinbase’s transaction
costs consist of account verification fees, payment processing fees, and fraud loss
expenses. As we show in Exhibit 54 and Exhibit 55, these have been a relatively
consistent share of volumes and revenues over time. We primarily model transaction
costs as a percentage of transaction revenues, where this level has historically
represented a low to mid teens share of transaction revenues over time. That said, we
believe institutional volumes are likely to carry a lower cost to execute than retail
volumes (more efficient payments, lower fraud cost, greater volume per account), so a
continued mix shift towards institutional volumes could lead to a continued downdraft in
transaction costs relative to volume over time.

'''Breakeven analysis: we see breakeven levels of profitability in a down 50% scenario relative to our estimates'''

While it is difficult to directly map changes in the level of crypto prices to COIN’s top line
given the volatility of prices as well as the fact that the absolute level is highly impactful
on volumes, in the following exhibit we attempt to illustrate roughly the sensitivity of the
model to prices relative to our base case. We use the historical relationships derived in
Section 2 to sensitize transaction revenues across various price and volatility
environments. To account for the potential drop-off in retail interest levels in a severe
down scenario, we include an additional 25% cut to MTUs in our down 50% scenarios
in addition to using the historical relationship. For expenses, we assume that sales and
marketing and transaction costs, as well as ~25% of technology costs will vary with
revenues while the remainder of the expense base is unaffected. This results in roughly
breakeven levels of profitability at the down 50% scenario, which would imply BTC
prices in the ~$20k range.

==Valuation: We see 36% upside to our $306 12 month price target==

We value COIN using a multiple of 2023E sales. In deriving our price target we looked at
a number of market structure and payments-related companies. While market structure
and brokerage comps make sense to us given COIN operates a cryptocurrency
exchange, we believe the long-term thesis in crypto revolves around the adoption of
crypto for commerce and payments, and thus we think the expansion of the crypto
ecosystem (and thus the prospects for Coinbase) are similarly tied to a payment
adoption curve over the long run. We believe looking at revenue multiples is appropriate
given the volatility of COIN’s operating margins (discussed in the last section), as well as
the company’s messaging that they are not optimizing the P&L for the current revenue
base and expect to operate at breakeven through the cycle. While we are unsure if we
will see these lower levels of profitability in the near term given the ongoing bull market
in crypto and COIN’s high levels of profitability, investors will likely turn to revenue
multiple in future periods of depressed profitability.

Our comp group includes MKTX, TW, CME, NDAQ, ICE, IBKR, SQ, and PYPL. These
peers currently trade at roughly 14x on average in terms of EV/sales on a 2022E basis.
The payments comps tend to trade a high teens revenue multiples whereas market
structure comps tend to trade at low teens multiples (Exhibit 60). Given our base case
for the crypto ecosystem to remain a largely store-of-value oriented ecosystem, we
primarily focus on the market structure comps for the purposes of valuing Coinbase, and
view the higher multiples for payments related names as representing an upside
opportunity should the crypto-native ecosystem develop further. To arrive at a 1 year
forward view, we apply a 13x multiple to our 2023 estimates. We then adjust for our 1
year forward net debt assumption of roughly $4.2bn in net cash, implying a 1 year
forward market cap of ~$81bn. This implies a 12 month target price of $306, implying
36% upside to shares from current levels. Given the sensitivity to crypto prices, we also
believe it is helpful to show how our valuation would change under various crypto
market cap assumptions. Thus, in Exhibit 59, we show the resulting valuation for a
100% increase in crypto prices as well as a 50% decline in crypto prices (these are the
same numbers derived in Exhibit 56). In a + 100% scenario we believe COIN’s valuation
could increase by as much as 70%, whereas we see roughly 30% downside in our
down 50% scenario.

==Regulatory concerns and key risks to our price target and investment
view==

The nascency of the cryptocurrency industry as well as its overlap with one of the core
functions of governments in providing a means of exchange in the economy lead to
significant risks to the crypto industry over time. Below we walk through a handful of
these risks.

Risks of governments moving against cryptocurrencies: One of the biggest risks to
our view is geopolitical, as governments usually have sole responsibility for
administering local currencies, and a product such as crypto that presents itself as a
substitute for government fiat currency could face additional regulation or outright bans
in the future. For example, in March 2021, the Indian government was in the process of
proposing an outright ban on cryptocurrencies that would prevent the ownership,
mining of, and transactions in cryptocurrencies. In addition, the Turkish Central Bank
recently decided to ban cryptocurrencies as a form of payment, citing concerns over
volatility and the lack of regulation. While we believe the nature of a decentralized
system running on the internet would make it very challenging to completely get rid of
cryptocurrencies, we believe the biggest lever that governments have at their disposal is
the control over the convertibility of fiat currencies into cryptocurrencies. If governments
restrict banks from facilitating the convertibility of cryptocurrencies with fiat currency by
not allowing banks to do business with crypto firms or to offer depository services to
crypto firms, we believe the value of crypto currencies would be negatively affected as
their perception by the public as a store of value or a medium of exchange would
diminish. This could potentially result in falling crypto prices, which in turn could lead to
accelerating pressure on crypto prices, as declining value could lead to further loss of
faith in crypto prices, not dissimilar to a bank run in a traditional banking sector where
fears over the convertibility of bank deposits into cash leads to withdrawals, leading to a
liquidity crunch and a self-fulfilling prophecy of total loss of faith of the bank’s ability to
remain solvent.

The current state of government regulation is mixed. The United States has recently
taken steps to normalize the crypto industry by allowing banks to offer custody for
crypto assets and by providing OCC charters to certain custody agents in the industry.
In China, the government has been experimenting with central bank digital currencies, a
move that has the potential to drive more crypto-native means of exchange activities
through state-backed crypto assets, which could be a catalyst for crypto adoption. That
being said, it remains unclear whether state-sponsored crypto assets will catalyze or
crowd out non-state-sponsored crypto assets over the long term.

Risks from US securities regulations: The regulatory status of various
cryptocurrencies has been in flux over time and could lead to uncertainty in the future if
the regulatory framework shifts. Today, Bitcoin and Ethereum, the two largest
cryptocurrencies are officially regulated as commodities in the United States and
therefore fall under the CFTC for regulation. That said, the regulatory status of other
crypto assets is less clear. For example, in December 2020, the SEC filed a lawsuit against Ripple Labs, the company behind the Cryptocurrency XRP, claiming that the
company conducted $1.3bn of unregistered securities sales, implicitly deeming XRP to
be a security. This led to a selloff in XRP (Exhibit 62 - although it has more than
recovered since then) and resulted in some cryptoexchanges, including Coinbase, to
delist XRP. Coinbase noted in their S-1 that they have a policy of not listing assets that
are likely to be deemed a security given that the company does not have the regulatory
licensing to list such assets. However, because of the lack of clarity around this issue, it
is possible that COIN could incorrectly categorize certain cryptocurrencies, as was
alleged to be the case with XRP. While this is the most recent example of such issues,
we believe that in general, the changing landscape for cryptocurrency regulation will be
a focal point among investors. 

KYC / AML: In the traditional finance space, banks and payment companies play a large
role in detecting and preventing money laundering, from reporting large cash
transactions to flagging suspicious activity, as well as conducting rigorous “Know Your
Customer” (KYC) checks when onboarding new clients. This, combined with the highly
ringfenced nature of the USD payments system (with a few exceptions, only banks can
hold and transfer Federal Reserve balances) means the US Dollar system and other
similar monetary systems have significant safeguards in place to prevent money
laundering and transfers of illicit funds. The decentralized nature of cryptocurrencies, on
the other hand, means there are fewer inherent checks on potentially illicit behavior, and
thus there is a very high bar to clear for cryptocurrency companies to accept fiat
currency payments and allow for deposits and withdrawals of fiat currency, because
there is a risk that the money being deposited or the cryptocurrency used to fund the
fiat currency withdrawals was sourced through extra-legal activities. Crypto is often
compared to “digital cash” due to it sharing some of the same features of physical cash,
such as being a bearer instrument and not requiring an intermediary to facilitate
transactions. The decentralized nature of physical cash without oversight from
intermediaries is why physical cash is often used for illicit activities, and the ability to
transfer cryptocurrency in large sums across great distances means that there is an
even greater risk of illicit activity in cryptocurrencies relative to cash. Thus, there is a
much higher bar for KYC/AML controls to ensure compliance. 

Coinbase is widely reported to have taken a measured and deliberate approach to
compliance, having worked with regulators from its beginnings to invest in the
infrastructure to satisfy BSA/AML laws and gain access to the US banking system. We
expect that as a ~$2tn asset class, governments around the world will continue to place
a high degree of scrutiny on compliance regimes at crypto-native companies, and the
larger the space becomes and the more regulatory focus it attracts, the greater
competitive moat COIN’s approach to compliance should represent.

Cybersecurity / custody risk: Many management teams at traditional financial services
companies will cite cybersecurity incidents as the risk to the business that “keeps them
up at night.” The cryptocurrency industry faces similar issues with the added risk that a
cybersecurity incident in which hackers gain access to the private keys of customer
crypto wallets can result in the permanent loss of customer funds. The immutability and
cryptographic properties of blockchain technology make it effectively impossible for the
funds to be recovered when the private keys are lost. While the same cybersecurity
concerns that exist at traditional financial institutions, such as personally identifiable
information, account numbers, and confidential information also exist for crypto-native
companies, a cybersecurity event has the potential to be the digital equivalent of a bank
robber gaining access to the bank vault, with the ability to instantaneously transport
away the contents. The most high-profile instance of this happening was in the 2014
collapse of Mt Gox, the then-largest crypto trading platform, which became insolvent
after the loss of BTC 850k. This issue highlights the importance of a strong custody
solution that safeguards assets.

Coinbase highlights their track record of having never had a cybersecurity incident that
resulted in the loss of customer funds, and we believe the strength of their custody
offering represents a key advantage over time, particularly for institutional clients.
However, a hack or loss of customer funds wold likely lead to negative impacts on the
broader industry, resulting from a loss of confidence in the safety of crypto assets, and
also potentially leading to lower crypto prices overall, which would indirectly have
ramifications for COIN earnings.

Lastly, we note that at a more operational level, downside risks to our positive
investment view on COIN include lower levels of crypto volatility and lower commission
rates.