What is DeFi?

In the past few months I’ve been to three conferences at which participates couldn’t agree on a definition of DeFi. 

Here’s an example, from a DeFi conference 😳

This sort of ambiguity won’t do, at least not in our context. 

Here at Graychain we talk about DeFi all the time. We analyze DeFi data, we build DeFi tools, and we publish reports that cover DeFi. Needless to say, we have a very clear understanding of what we mean by DeFi!

For us, 

DeFi includes any financial service that is not designed to require a central authority. 

That definition gives us a lot of latitude, but easily excludes any traditional financial that happens to process digital assets. In other words, we adhere to the “small tent” definition of DeFi, while acknowledging that true De(centralization) is aspirational. 

If DeFi is a spectrum, how do we decide when something is NOT DeFi?

  • If a service is regulated, it’s not DeFi.
  • If a service is not managed by a smart contract, it’s not DeFi. 

Simple. Now let’s look at some examples.

If Compound DeFi?

Yes. It’s globally accessible, doesn’t KYC, and can’t be shut down by any state regulator. The Compound service is encoded in smart contracts.

Is BlockFi Interest Account (BIA) DeFi?

No. BlockFi is regulated. 

Is Binance DEX DeFi?

Yes. Although Binance itself is regulated, its DEX isn’t

We are clearly focused on credit, which means that we work with projects / companies in the lending space. The biggest DeFi lenders are:

  • Compound,
  • Maker,
  • dY/dX, and 
  • nüo.

None of them are completely De(centralized), but they aspire to be, which is all we can expect at this point in history.

The First Crypto Credit Industry Report

We just published the first crypto credit industry report. Every issue will include a synthesis of all the industry data that we think matters, as well as our commentary on that data.

While putting this first report together we found that the industry is growing at 20% MoM. We arrived at this number starting with two completely different metrics. Any industry growing that fast should be of interest to the investment community.

Researching and writing the report isn’t charity work. We want to sell our data to every crypto lender. Putting this report together allows us to deeply understand what they need. 

There’s a lot of data we didn’t include. That was because we either didn’t trust it or didn’t have time to talk to all the vendors. 

As this article points out, our report shows that industry profit margins are low. There are a few reasons for that. 

  1. Some of the lending platforms are not designed to generate profits in a traditional way, so the profit we can attribute to them is $0.
  2. We heavily discounted some vendor claims because we couldn’t make the numbers add up. We may have been too conservative. Or not.
  3. It’s a brand new industry, 24 months old, and we’re seeing lots of experimentation with new business models. Although the traditional fintech crowd is focused on profit, the more innovative crypto crowd is more focused on utility. From their perspective, direct profits either don’t matter or will come later.

Our biggest takeaway from this work is how quickly things are evolving. Having spoken to so many people in the industry, we are convinced that this rate of growth will continue, and may even accelerate in the short term. A lot of new vendors and products that should soon see the light of day are being developed.

We’re definitely looking forward to the next report!

What will Happen in the DeFi Lending Area: Credit in Crytpo and its Huge Market Potential

Coin Tokyo Article

If you read “Traditional Credit – A Primer” you know all about how traditional credit works. You don’t actually, but you know enough to take a plunge into the increasingly-strange world of crypto credit.

The Distant Past

Once upon a time (3 years ago), very generous VCs and ICO investors gave some enterprising startups money to start unsecured lending businesses. Some, like BTCJam went out of business,while others, like Ripio changed their business model . These startups tried all sorts of clever ways to make sure people paid back their loans. Most of them didn’t. The investors’ money disappeared. 

The Recent Past

From the ashes of these early failed experiments rose today’s DeFi Lenders. You’ve probably heard of some of them:  SALT, Compound, Dharma, dxdy, Cred

These guys are doing pioneering work. They’re allowing crypto holders to leverage those assets to borrow more crypto or fiat (via stablecoins). This is pretty great if you’re convinced your crypto is going to go up in value, or just need cash without incurring a capital gain.

Although collateralized lending is not new, these companies are introducing all sorts of innovations into the process. This is thanks to the nature of crypto. Because of its recent volatility, stablecoins are often part of the equation. Because many of the transactions are managed by smart contracts that can react to data in real-time, contract parameters can be more flexible than they are traditionally. Since blockchains aren’t shut off on weekends, contracts can liquidate or settle at any time.

Despite the innovation, the companies building and managing these loan platforms are taking almost no risk. If the value of crypto looks like it will drop below the value of the outstanding loan, or if the borrower fails to meet the repayment terms, the lender liquidates the collateral and still makes a profit. 

Unfortunately, there’s a downside to not taking any risk. These companies are leaving money on the table.


Today there’s a lot of planning going on, and it’s exciting.

The current batch of DeFi lenders are working to reduce the amount of collateral they have to hold and to increase the amount of time they can comfortably wait before liquidating a loan. Both of those will increase their profit margins. 

They’re also working on increasing their exposure to lenders’ behavior, thereby requiring even less crypto as collateral. They will replace crypto with reputation. This will be possible thanks to the availability of crypto credit data. As a result crypto lenders will enjoy the same dynamics as non-crypto lenders who rely on both hard assets and credit bureaus to mitigate the risk of delinquent borrowers.

But today’s DeFi lenders aren’t alone. Lots of companies are quietly planning their move into crypto lending.

The Near Future

Anyone with a background in financial services knows how critical a healthy credit market is to the whole financial system. Most of the people in today’s DeFi space have that background.

Anyone holding digital assets is going to want to lend them out. That’s how traditional banks make money, and that’s how digital custodians will make money. 

Crypto credit data is going to allow a flood of new entrants into the lending space.

Today, crypto lending is primarily a trading strategy. People leveraging their assets in this way are very sophisticated. The use of crypto for day-to-day transactions is growing and will continue to grow. In parallel we will see a demand for more traditional loans, both partially-collateralized and uncollateralized. The supply for this demand will be waiting. Volumes will surprise everyone.

The (slightly less) Near Future

One very interesting property of crypto credit data is that it’s associated with blockchain addresses, not people. This has two unexpected implications.

First, it means that loans can be made anonymously. If an address has a positive credit score, does it matter if we know who’s behind it? Absolutely not. This type of lending won’t be possible in a lot of jurisdictions, but the world’s a big place, and you can bet your bottom dollar that a lot of lenders in the future will be jurisdictionless. Some will be completely autonomous.

And speaking of autonomous, besides a person’s wallet, what else can you find behind a blockchain address? A smart contract. Who’s to say that smart contracts won’t need to borrow money from time to time? If a contract has a good reputation, shouldn’t it be able to? No such contract exists yet, but it’s easy to imagine a dapp in a supply chain needing short-term liquidity to bridge a transaction.

I did mention that things were going to get strange, didn’t I?

From Zero to One (Trillion)

Today, the traditional financial system issues trillions of dollars of loans every year.

Dozens of companies in the crypto space are in the process of building out the infrastructure required to issue loans on a similar scale. That would be exciting enough, but the story won’t end there. Thanks to the unique properties of cryptocurrencies, we are going to see the emergence of never-before-seen credit products that will support use-cases we can’t even imagine today. 

One trillion is a certainly a low estimate.

About Graychain

Graychain is the world’s first crypto credit bureau. It was founded in 2018 after the founders built a peer-to-peer lending platform that they never launched. They never launched it because they needed credit data, and it didn’t exist.

Graychain is based in Hong Kong with personnel in the US, South-East Asia, Europe, and Africa.

About Paul Murphy

Mr. Murphy is one of Graychain’s founders and its CEO. His career in the software industry has spanned over twenty years and four continents.  

The first ten years were dedicated to understanding and building large systems on Wall Street for clients like J.P. Morgan and Salomon Brothers. Mr. Murphy’s work in this area allowed him to explore a broad range of computing solutions.

After 9/11, Mr. Murphy moved to London to work at Adeptra, a pioneer in the use of automated outbound calling in the area of credit card fraud detection and prevention. The systems Mr. Murphy developed made extensive use of text-to-speech and voice recognition technologies.

Exploring the intersection of telephony and computing eventually led Mr. Murphy to found Clarify, a pioneer in speech recognition and natural language processing, both of which make extensive use of AI.

Graychain is Mr. Murphy’s fifth startup. He firmly believes that cryptocurrency-based financial systems will transform the world in ways that will surprise, frighten, and delight us all.

What will Happen in the DeFi lending area, A Primer: Traditional Credit and its Importance

Coin Tokyo Article

Why Credit Matters

Credit is the most boring topic in finance. It’s also the most exciting. 

Credit is boring when it exists, invisibly, as part of the fabric of a financial system. In places where this is the case – like Japan – only the absence of credit would be exciting, exciting the way natural disasters are exciting. In places like Japan and the US, credit is institutionalized.

In other parts of the world, like many countries in Africa, Latin America, and South-East Asia, institutional credit barely exists. Sometimes it doesn’t exist at all. People there can’t buy homes or cars on credit and they can’t take out loans to start a business. Countries without institutional credit, however, tend to have very well-developed informal (or social) credit systems. In India, it’s not unusual to tell a shopkeeper that you’ll pay him tomorrow for the milk you are bringing home today. Imagine doing that at a 7-11 in Tokyo.

All credit removes friction from people’s lives. Informal credit allows people to buy milk even if they don’t have enough money on hand. Institutional credit allows people to buy a car even if they don’t have enough money on hand.

Credit is a critical lubricant in any financial or social system.

But what is Credit?

Credit is Trust.

The shopkeeper in India knows the person who has taken his milk without paying. The shopkeeper (lender) knows that the borrower lives in the neighborhood, he knows how the borrower has behaved well in the past, and he knows the borrower’s friends and acquaintances. These social connections make it highly likely that the borrower will eventually pay for the milk. This “social credit” guarantees the loan.

When credit is institutionalized, that social element doesn’t exist in that way. Even if your banker knows you, the bank’s shareholders don’t. How then can they trust that you will repay a loan without a social guarantee? Very simply, in exchange for lending you money, they hold on to something you don’t want to lose, either an asset, your reputation, or both. The asset – which we can call collateral in this case – is given to secure the repayment of the loan. In English, that means that if the loan isn’t repaid, the borrower gets to keep the asset. If you borrow money to buy a house, the bank gets to keep the house if you don’t make contracted payments. This is called the liquidation of a loan. The collateral is sold for cash to recover the money lent. 

Collateral therefore acts as an enforcement mechanism for the repayment of the debt. Simple, but, unfortunately, not enough in a lot of cases.

What happens if the property market crashes between the time you take out a loan to buy a house and the time the bank realizes that you aren’t going to repay it? They may then be unable to liquidate the house for anywhere near the amount of money they lent you to buy it. 

So what else can an institutional lender get from a borrower to secure a loan? Something very similar to what the shopkeeper in India has: the borrower’s reputation.

Reputation as Collateral

Reputation is precious, and it is most precious to its owner. Its loss can be devastating.

The shopkeeper can destroy someone’s reputation by telling the community that one of its members doesn’t honor their debts. Once that’s known, the individual in question is unlikely to be offered credit by other merchants.

Institutional lenders, without the social network, essentially have the same power. They have it thanks to the credit reporting and scoring systems that exist in every developed economy. 

The mechanism is simple:

  1. Lenders report their clients’ behavior to credit bureaus.
  2. When an institution is asked to extend credit, they ask the credit bureau for a summary of the potential borrower’s past behavior. 
  3. The credit bureau summarizes that behavior into a credit score. 

Someone with a low credit score is unlikely to:

  • get another loan, 
  • be given a post-pay phone contract, 
  • be able to rent a home, and even
  • be able to get certain jobs. 

Loss of reputation in the form of a low credit score can be devastating.

This is how credit works traditionally, but what happens in the crypto space?
For the answer to that question, read “Crypto Credit – From Zero to One Trillion”.

Crypto Credit in Nairobi

On May 23rd, from 18:00 to 19:30, I will be at Utu House in Nairobi. I’ll be talking about credit scoring, cryptocurrencies, and the very strange future that many of us are sleepwalking into.  Some of the topics I’ll be covering are discussed in this recent blog post.

If you’re in Nairobi, please join us at:

Utu House
Lower Kabete

If you have any trouble, please call 0715 656755.

NB: Utu House does not provide parking so best to use taxi services such as MARAMOJA transport.

Credit Where Credit is Due

Very few places on the planet enjoy a healthy consumer credit environment, an environment in which people are given the opportunity to prove they can manage debt. Where it exists, magic happens. People in those places become an important part of the economic engine that drives a country forward.

The US enjoys a particularly healthy credit environment. People there can earn credit and take it from one institution to another. They can, for example, take out a loan from one bank, pay it back, and then use the resulting positive reputation to take out a bigger loan from another bank. There aren’t many places in the world where this is possible.

In countries in which consumer credit is otherwise healthy, credit is often tied to a lender. In France, for example, I can build up credit at one bank, but that will mean nothing if I change banks. This is OK, but far from ideal.

Both the US and France are rich countries with well-developed financial systems. But what about places in which few people even participate in the financial system? In Argentina, for example, 80% of people don’t. If they have a bank account, they don’t keep money in it, and they don’t have credit cards or bank loans.

This lack of consumer credit, which affects most people on the planet, is harmful to both individuals and the countries they live in.

I won’t go into how we got here. It’s a long, complicated story. Instead I’d like to focus on what people are doing about it.

Most governments are doing nothing to fix this problem. This will surprise no one!

Traditional lenders, banks primarily, are also doing very little. This isn’t because they wouldn’t love to solve the problem — most bank revenues are derived from loans! — but because of economic and political barriers such as:

  • poverty,
  • corruption,
  • poor education, or even
  • lack of government support.

It turns out that bootstrapping a credit system is very difficult! This leaves the door open to scrappy innovators.

Before I talk about these innovators, I need to make one other thing clear:

You can’t lend people money without (1) having a way to make sure you’ll get the money back, (2) having a reasonably good idea that the borrower is likely to repay the debt , or (3) both.

To make sure we get our money back, we need some kind of Enforcement. Physical enforcement is the primary mechanism behind criminal credit. In Italy, for example, someone who might not be able to get a loan from a bank, could easily get one from the local mafia. If the loan isn’t repaid as agreed, the mafia might enforce the “contract” by physically harming the borrower. This is effective, but also rather unsavory. Thankfully, we’ve developed a less gruesome enforcement mechanism. If someone fails to repay their debt, we harm their credit score instead of their person. In a well-functioning financial system, a person’s credit score will affect their future ability to borrow, so keeping it healthy is important!

But what about the second component: being able to predict whether or not someone is likely to repay a debt? It turns out that a person’s credit score is derived from data about their past behavior, so it should help us predict future behavior.

A credit score is therefore a key component of our enforcement mechanism and our prediction mechanism. Very clever!

This score works really, really well in a traditional financial system because all incentives are aligned behind it. But what about outside that context? In that messy world, we need other ways to predict a borrower’s behavior.

This is where innovative companies like Utu come in. They use social media, gaming, and transport data to determine whether or not someone might be a good credit risk. Utu use this data to produce credit scores that allow lenders to make smart decisions. This is difficult work! Collecting the data is difficult. Predicting people’s behavior from that data is even more difficult. But the more lenders use the data and provide feedback about the outcome of loans, the better the predictions become. It won’t be long before your social profile will be a better predictor of your creditworthiness than credit history.

This is an astonishing development. Twenty years ago, no one would have predicted that this was even remotely possible. These alternative credit scores will change the economic environment in countries where most unbanked people live today.

Careful readers will have noticed that these alternative credit scores are not exactly like their traditional counterpart. They cannot be affected by failure to repay a loan since that data is not used to compute the score. That data is not used because transactions between lenders and borrowers are private, and therefore invisible. An alternative credit score, unlike its traditional counterpart, cannot be part of an enforcement mechanism, with one exception.

If loans are managed on public blockchains, transactions are public, so they can be taken into account when computing alternative credit scores. Is that happening? It is.

Cryptocurrencies are fast becoming the foundation of a new financial system, especially in places without a strong existing infrastructure.

This new financial system has properties that don’t exist in its traditional counterpart. In other words, we aren’t witnessing the emergence of a “more efficient” system, we’re witnessing the emergence of a different system. Anonymity and the ability to transact peer-to-peer change the dynamics completely. In the credit space, these different dynamics are clear. We therefore need different ways of predicting and enforcing good behavior.

Companies like Utu are a big part of this story since they provide alternative behavioral data. But we need more than that in the crypto space. We need companies that approach the credit data problem differently than traditional credit bureaus. These companies need to consider anonymity, pseudonymity, and the global, trans-national, nature of blockchains.

Graychain is doing just that. We are producing credit scores for blockchain addresses, not people, and we are working with data providers like Utu to understand the behavior of people behind these addresses. Our work is providing the foundation for a new generation of lenders, lenders that will be critical to this new financial system.

Getting Started with Crypto Wallets





  1. Go to your browser’s extension page (Chrome, Brave, Firefox, Opera).
  2. Search for “metamask”.
  3. Install it.
  4. Click on the fox in the extensions area of your browser.
  5. Accept the terms of service.
  6. Create a password to protect your wallet (they call it a Den).  Make it amazing.
  7. STORE THE 12 WORDS YOU NEED TO RECREATE YOUR ACCOUNT. Please assume your computer will be stolen and that you’ll forget or lose the amazing password you generated in step 6. Those 12 words are the only thing that will allow you access to your money under those circumstances!


You now have a wallet in which you can store your ETH!

If you have an account on an exchange, you can now send money to your wallet from your USD, GBP, etc. bank account.

If you don’t have an account on an exchange, read our handy Guide to Crypto Exchanges.

If you’re impatient or don’t plan on opening an account on an exchange:

  1. Find your wallet address.
  2. Give it to a friend who has Ether.
  3. Ask him/her to send you some.  

Voila! You now have a working wallet with some money in it.  Now run Graypay, and get started!



If you’re new to cryptocurrencies, you should put this guide aside until you’re ready to learn more.  If not, read on!

MetaMask is a “hot wallet”, i.e., it’s online whenever your computer/browser is online. There are two other types of wallets you should know about, both of which will work with MetaMask:



Hardware wallets are also called “cold wallets”. They are hardware devices that are not (or hardly ever) connected to the Internet, so are extraordinarily difficult to hack. When you need to add/remove money from a hardware wallet, you connect it to your computer and confirm transactions via the device itself. It’s slightly magical the first few times you do it!

MetaMask currently supports hardware wallets from:

Check your version of MetaMask to find out what hardware wallets it supports.


These can be hot or cold.  They are wallets that are kept by a custodian, typically, an exchange.

If you open an account on the Coinbase exchange, for example, they will create a wallet for you and give you access to it via their account credentials (e.g., username and password).  It’s completely transparent and therefore convenient. Most exchange users don’t even realize that the exchange is their wallet custodian.

Some custodians, like Xapo, keep a copy of your wallet in cold storage in a Swiss mountain (that’s not a joke).  It’s extremely safe, but very slow. Definitely not the right solution for Graypay users (!), so we’re going to focus on hot custodial wallets.

Remember those 12 words that were generated when we created our MetaMask wallet? When you create a hot wallet with a custodian, you also get 12 words! Handy. It turns out that you can use those 12 words to import (or access) your custodied wallet from within MetaMask.  All you need to do is “Import using account seed phrase” instead of logging in.


About us

  • Our team
  • Our story
  • Our partners

© 2019 Graychain Ltd. All rights reserved. Terms of Use & Privacy Statement