Deciding how to spend your blockchain treasury

Blockchain treasuries should embrace a diversity of mechanisms including grants, tenders and prizes.

By Darcy Allen, Chris Berg, Sinclair Davidson and Jason Potts. Originally posted at Cryptoeconomics Australia.


While multi-billions of dollars of value currently sit within blockchain community treasuries, there is no clear consensus on how to spend them.

Recent proposals to spend defi treasuries include flashy governance mechanisms (e.g., Gitcoin’s quadratic funding, or retroactive public goods funding), controversial allocation to third parties (e.g., $20 million from UniSwap’s treasury allocated to the Defi Development Fund), and selling portions of the treasury in the name of diversification (e.g., the current SushiSwap proposal).

The destination of treasury funds also varies widely from funding a specific protocol upgrade, to investments in highly uncertain long-term scaling projects. Likewise, there is no single solution or mechanism to solve the treasury spending problem. And so in this post our aim is to provide some guidance in how to spend a blockchain treasury, with inspiration from existing approaches to procure innovation elsewhere.

Many mechanisms have been developed and deployed across the public and private sector to fund research and development. These include grant programs, prize pools, and tenders, that help to procure innovation and other services in different contexts. The custodians of defi treasuries would do well to consider the advantages and disadvantages of these approaches, integrating them into their designs.

Your treasury spending problem

Before deciding to spend your treasury your first task is to make sure that it is robust to opportunism. Your treasury must be robust to hacking, theft and attacks (both from insiders and outsiders). This is the priority given the important signaling effect of a large, secure and public treasury.

But what then? How should you spend it?

Here you must understand the precise problem that you face. No one knows (or could know) how to optimally deploy treasury funds. Your problem is an entrepreneurial one. You must deploy a scarce resource in a dynamic and competitive environment without full information. In this way your problem is somewhat similar to that of a venture capitalist or philanthropic foundation.

While the choice of an entrepreneur might be guided by profit maximisation, your choices must be guided by the objectives set by your community: ensuring security, ecosystem growth, community development and innovative projects. In some ways this means your challenge is subject to even greater uncertainty than a profit-seeking entrepreneur (who has a clear metric — profit — by which to recalculate).

But all is not lost. Each spending decision doesn’t have to be guesswork mired in controversy. What you need are some mechanisms to discover who to fund, on what projects, at what time, under what conditions, and in what priority.

Designed well, those mechanisms ameliorate uncertainty by putting to use distributed information about what should be done. They should align incentives between ecosystem participants and treasury objectives. And they should aid in coordination and matching between different actors in solving ecosystem problems.

In most cases, spending a treasury is not simply about writing and enforcing contracts to produce known goods and services. It is an ongoing process of learning, discovery and feedback. The design of treasury spending must be designed to understand that spending a treasury is a dynamic process: the challenges that a particular ecosystem faces today will likely different in only a few months time, shifting the nature of the entrepreneurial problem.

We recently proposed a new design for treasury grants ecosystems. Each budget period, tokenholders fund a small number of philanthropists to run their own grant programs (each deciding how to achieve the overall treasury or round objectives). At the end of each period the tokenholders rank those philanthropists, and they compensate them based on that rank. Such a design imparts some of the entrepreneurial and competitive processes needed to navigate your entrepreneurial problem.

But grants are just one tool in a blockchain treasury toolkit. And so in the next section we outline how to deploy different funding mechanisms depending on two fundamental problems: a knowledge problem and a coordination problem.

Prizes, tenders and grants

Here we explore how to think about deploying three mechanisms to procure your innovation activities: grants, prizes and tenders. To decide between these options, or simply whether your team should do the project internally, you need to think about two questions.

The first question is whether you know the problem that you’re trying to solve. Do you know what problem you need to spend the funds on? The answer is ‘yes’ if you can pinpoint precisely what you’re trying to solve — such as a protocol fix or a specific wallet integration. But the most common answer to this question, at least at this point in defi development, is ‘no’ — and so you face what we can call a knowledge problem.

The second question is whether you know how to solve the problem. Remember, your problem is an entrepreneurial one: you don’t just need to know what to do, but also access to the combination of resources (capital and labour), and in the necessary order and quantities, to solve that problem. If you don’t know how to solve the problem — including whether you can’t identify the problem at all — then you face what we can call a coordination problem.

Why does answering all of these questions matter? Because they suggest a different approach to producing the goods and services (and innovation) that your blockchain ecosystem needs. Here are your choices:

  • Just do it. If you don’t have a knowledge or a coordination problem then your choice is easy. You know what you want and roughly how to do it. This means that rather than buying the thing over a market, you should just make it in-house. If it’s an ongoing problem, employ the best person and pay them a wage. Sure, there are possible incentive and scale problems here, but those costs might be less than going to market.

  • Prize. On the opposite end of the spectrum, you don’t know precisely what you want and you don’t know how to do it. That means you have both knowledge and coordination problems. Here your best option may be a prize. You will fund innovation after they have been developed, perhaps through a committee decision or a community vote. You might not be willing to fund your moonshots beforehand, but you’re willing to fund them after the fact. Your prizes might be challenge-based (e.g., scaling metrics) as an effort to spur competititve dynamics between innovators around the broad goal you have.

  • Tender. If you know what problem you are facing, but you don’t know precisely how to do it, you are dealing solely with a coordination problem. Use a tender. You should specifically define the problem you need solved, or the product you need developed, and seek for competitive applications from a broad range of applicants. For tenders to be effective you need to reasonably be able to assess the merits of how the various proposed approaches will solve your problem. Unfortunately most defi treasury challenges are not like this.

  • Grant. If you don’t know what you want (you face a knowledge problem), but you believe that the resources within your community are best deployed to solve that problem, then you should consider a grants program. Many defi ecosystems have implemented grants programs. Of course if you don’t know what you want (a knowledge problem) it is difficult to know how to do it (avoiding a coordination problem), but the focus here is on if you will know it when you see it. You want your community to pitch you. That grants program might be themed, or have various levels, depending on the characteristics you’re seeking to achieve.

Your choices are not perfect and their boundaries are blurry. Hack-a-thons, for instance, may also be deployed as an innovative process, and might either be structured like grants programs or as prizes.

Nevertheless, what we have provided here is a schema for choosing between treasury spending mechanisms. Importantly, this is not a one-size-fits-all approach to treasury spending (e.g., a DAO vote), but rather understands that different problems need different institutional solutions. And those solutions must emphasise how to solve the entrepreneurial challenge at the heart of treasury spending.