The book is now published at:

https://www.amazon.com/DeFi-Future-Finance-Campbell-Harvey/dp/1119836018/ref=sr_1_3

The following is an older, unedited version of the manuscript

DeFi and the Future of Finance*

Campbell R. Harvey

Duke University, Durham, NC USA 27708

National Bureau of Economic Research, Cambridge MA USA 02138

Ashwin Ramachandran

Dragonfly Capital

Joey Santoro

Fei Protocol

ABSTRACT

Our legacy financial infrastructure has both limited growth opportunities and contributed to the inequality of opportunities. Around the world, 1.7 billion are unbanked. Small businesses, even those with a banking relationship, often must rely on high-cost financing, such as credit cards, because traditional banking excludes them from loan financing. High costs also impact retailers who lose 3% on every credit card sales transaction. These total costs for small businesses are enormous by any metric. The result is less investment and decreased economic growth. Decentralized finance, or DeFi, poses a challenge to the current system and offers a number of potential solutions to the problems inherent in the traditional financial infrastructure. While there are many fintech initiatives, we argue that the ones that embrace the current banking infrastructure are likely to be fleeting. We argue those initiatives that use decentralized methods - in particular blockchain technology - have the best chance to define the future of finance.

Comments: Please comment directly on the live version of our paper. It is available here.

Keywords: Decentralized finance, DeFi; Fintech, Flash loans, Flash swaps, Automatic Market Maker, DEX, Decentralized Exchange, Cryptocurrency, Uniswap, MakerDAO, Compound, Ethereum, Aave, Yield protocol, ERC-20, Initial DeFi Offering, dYdX, Synthetix, Keeper, Set protocol, Yield farming.

JEL: A10, B10, D40, E44, F30, F60, G10, G21, G23, G51, I10, K10, L14, M10,O16, O33, O40, P10, C63, C70, D83, D85

*Current version: April 5,2021. We appreciate the comments of Dan Robinson, Stani Kulechov, John Mattox, Andreas Park, Chen Feng, Can Gurel, Jeffrey Hoopes, Brian Bernert, Marc Toledo, Marcel Smeets, Ron Nicol, Daniel Liebau Giancarlo Bertocco, Josh Chen, Lawrence Diao, Deepanshu, Louis Gagnon, Herve Tourpe, Vishal Kumar, Jullian Villella, Yash Patil, Manmit Singh, on an earlier draft. Lucy Pless created the graphics and Kay Jaitly provided editorial assistance.

________________

Table of Contents

1. Introduction 4

2. The Origins of Modern Decentralized Finance 6

2.1 A Brief History of Finance 6

2.2 Fintech 6

2.3 Bitcoin and Cryptocurrency 7

2.4 Ethereum and DeFi 9

3. DeFi Infrastructure 9

3.1 Blockchain 10

3.2 Cryptocurrency 10

3.3 The Smart Contract Platform 11

3.4 Oracles 12

3.5 Stablecoins 13

3.6 Decentralized Applications 14

4. DeFi Primitives 14

4.1 Transactions 15

4.2 Fungible Tokens 16

4.2.1 Equity Token 17

4.2.2 Utility Tokens 17

4.2.3 Governance Tokens 18

4.3 Nonfungible Tokens 19

4.3.1 NFT Standard 19

4.3.2 Multi-Token Standard 19

4.4 Custody 20

4.5 Supply Adjustment 20

4.5.1 Burn - Reduce Supply 20

4.5.2 Mint - Increase Supply 21

4.5.3. Bonding Curve - Pricing Supply 22

4.6 Incentives 22

4.6.1 Staking Rewards 23

4.6.2 Slashing (Staking Penalties) 23

4.6.3 Direct Rewards and Keepers 24

4.6.4 Fees 24

4.7 Swap 25

4.7.1 Order Book Matching 25

4.7.2 Automated Market Makers (AMMs) 25

4.8 Collateralized Loans 27

4.9 Flash Loan (Uncollateralized Loan) 28

5. Problems DeFi Solves 28

5.1 Inefficiency 29

5.1.1 Keepers 29

5.1.2 Forking 29

5.2 Limited Access 30

5.2.1 Yield Farming 30

5.2.1 Initial DeFi Offering 30

5.3 Opacity 31

5.3.1 Smart Contracts 31

5.4 Centralized Control 32

5.4.1 Decentralized Autonomous Organization 32

5.5 Lack of Interoperability 32

5.5.1 Tokenization 33

5.5.2 Networked Liquidity 34

6. DeFi Deep Dive 34

6.1 Credit/Lending 34

6.1.1 MakerDAO 34

6.1.2 Compound 39

6.1.3 Aave 46

6.2 Decentralized Exchange 49

6.2.1 Uniswap 49

6.3 Derivatives 55

6.3.1 Yield Protocol 55

6.3.2 dYdX 58

6.3.3 Synthetix 63

6.4 Tokenization 65

6.4.1 Set Protocol 66

6.4.2 wBTC 67

7. Risks[a] 68[b]

7.1 Smart-Contract Risk 68

7.2 Governance Risk 70

7.3 Oracle Risk 71

7.4 Scaling Risk 72

7.5 DEX Risk 73

7.6 Custodial Risk 74

7.7 Regulatory Risk 75

8. Conclusions: The Losers and the Winners 76

________________

1. Introduction

We have come full circle. The earliest form of market exchange was peer to peer, also known as barter. Barter was highly inefficient because supply and demand had to be exactly matched between peers. To solve the matching problem, money was introduced as a medium of exchange and store of value. Initial types of money were not centralized. Agents accepted any number of items such as stones or shells in exchange for goods. Eventually, specie money emerged, a form in which the currency had tangible value. Today, we have non-collateralized (fiat) currency controlled by central banks. Whereas the form of money has changed over time, the basic infrastructure of financial institutions has not changed.

However, the scaffolding is emerging for a historic disruption of our current financial infrastructure. DeFi or decentralized finance seeks to build and combine open-source financial building blocks into sophisticated products with minimized friction and maximized value to users using blockchain technology. Given it costs no more to provide services to a customer with $100 or $100 million in assets, we believe that DeFi will replace all meaningful centralized financial infrastructure in the future[c][d]. This is a technology of inclusion whereby anyone can pay the flat fee to use and benefit from the innovations of DeFi.

DeFi is fundamentally a competitive marketplace of decentralized financial applications that function as various financial “primitives” such as exchange, save, lend, and tokenize. These applications benefit from the network effects of combining and recombining DeFi products and attracting increasingly more market share from the traditional financial ecosystem.

Our book details the problems that DeFi solves: centralized control, limited access, inefficiency, lack of interoperability, and opacity. We then describe the current and rapidly growing DeFi landscape, and present a vision of the future opportunities that DeFi unlocks. Let’s begin with the problems:

* Five Key Problems of Centralized Financial Systems

For centuries, we have lived in a world of centralized finance. Central banks control the money supply. Financial trading is largely done via intermediaries. Borrowing and lending is conducted through traditional banking institutions. In the last few years, however, considerable progress has been made on a much different model - decentralized finance or DeFi. In this framework, peers interact with peers via a common ledger that is not controlled by any centralized organization. DeFi offers considerable potential for solving the five key problems associated with centralized finance:

Centralized control.[e] Centralization has many layers. Most consumers and businesses deal with a single, localized bank. The bank controls rates and fees. Switching is possible, but it can be costly. Further, the US banking system is highly concentrated. The four largest banks have a 44% share of insured deposits compared to 15% in 1984.[1] Interestingly, the US banking system is less concentrated than other countries, such as the United Kingdom and Canada. In a centralized banking system, a single centralized entity attempts to set short-term interest rates and to influence the rate of inflation. The centralization phenomenon does not just pertain to the legacy financial sector. Relatively new tech players dominate certain industries, for example, Amazon (retail) and Facebook/Google (digital advertising).

Limited access. Today, 1.7 billion people are unbanked making it very challenging for them to obtain loans and to operate in the world of internet commerce. Further, many consumers must resort to pay-day lending operations to cover liquidity shortfalls. Being banked, however, does not guarantee access. For example, a bank may not want to bother with the small loan that a new business requires and the bank may suggest a credit card loan. The credit card could have a borrowing rate well above 20% per year, a high hurdle rate for finding profitable investment projects. In developing countries, like India, this rate is closer to 40%.

Inefficiency. A centralized financial system has many inefficiencies. Perhaps the most egregious example is the credit card interchange rate that causes consumers and small businesses to lose up to 3% of a transaction’s value with every swipe due to the payment network oligopoly’s pricing power[f][g][h][i][j][k]. Remittance fees are 5-7%. Another example is the two days it takes to “settle” a stock transaction (officially transfer ownership). In the internet age, this seems utterly implausible. Other inefficiencies include: costly (and slow) transfer of funds, direct and indirect brokerage fees, lack of security, and the inability to conduct microtransactions. Many of these inefficiencies are not obvious to users. In the current banking system, deposit interest rates remain very low and loan rates high because banks need to cover their bricks-and-mortar costs. A similar issue aris