It was the best of times, it was the worst of times… I’m not saying that software engineers working on decentralized finance are the modern day Jacobins, but it does feel as if crypto has become completely politicized. (”Elizabeth Warren is Building an Anti-crypto Army”, whatever that means.)

I believe there are good faith debates to be had on the merits of open source, decentralized financial systems versus traditional, centralized systems. But it seems that those loudest in the anti-crypto faction (including policymakers and journalists with huge amounts of influence) remain confused about the basic use cases of DeFi, crypto, and financial technology in general.

I felt compelled to share a couple of recent anecdotes to help illustrate the current reality of financial products and services.

In a ‘Spring Cleaning’ mindset, I wanted to take a closer look at my personal finances. My initial goal was to build a database/spreadsheet of my account balances and transactions across TradFi and DeFi. There is absolutely no good reason why I should have to pull these numbers manually.

Setting up automation for DeFi was trivial. There are excellent, free APIs and applications that retrieve this data. For my TradFi accounts, I reluctantly signed up for Plaid. Ideally, I wouldn’t need to pay Plaid and provide them with my information, but there seemed to be no viable alternative for programmatically retrieving my own data.

After spending a while setting up Plaid’s authorization flow in a custom app I built, I attempted to connect a Vanguard brokerage account. Apparently, my brokerage account is not eligible for use with Plaid. In search of (seemingly non-existent) documentation of this limitation, I discovered that Plaid does have a “known issue” with Vanguard, though I believe it’s unrelated:

To maintain system stability, Vanguard currently limits access during high-volume windows. As a result, please expect unavailability between 9-11am & 3-4:30pm ET, and limited availability between 8-9am & 4:30-8pm ET. We recommend end users link Vanguard accounts between 5pm - 9am ET.

Seriously?

The transactions related to my credit card, which I opened through my bank, were also inaccessible via Plaid. It became apparent that just moving all of my TradFi accounts to a different bank would be a more expedient path.

Following lengthy in-person meetings, navigating complicated credit card product comparison charts, and signing hundreds of pages of PDFs on DocuSign (which I didn’t thoroughly review), I appeared to have an arrangement suited to my needs. I used the bank’s web app to initiate a transaction moving in some funds.

The transaction failed. I missed a call from an unknown number to authorize the transaction (which I wasn’t notified I should expect). It’s still unclear to me if the sending or receiving institution cancelled it. Ultimately, I brought a paper check to the bank and waited a few days for the transaction to complete.

Also, the application for the credit card that I was upsold on was denied. Then it was approved the following week. I’m not entirely sure what of my information these decisions were based on—or why the sudden change of heart—but, frankly, I’ve lost all interest in finding out.

To be extra explicit here: in my experience, the most basic use cases for financial technology (including account creation, retrieval of account balances, and transfer of funds) were far from smooth while using what is supposedly the best that the traditional system has to offer.

By contrast, I’m part of a team that recently received a grant for open source software we’re developing. I was notified that we could retrieve the funds on-chain and transferred them to my wallet. We created a new wallet to be used for the project and I transferred the funds there. We exchanged the tokens into ETH using Uniswap, bridged it to a different blockchain using Hop, exchanged some of the ETH into USD-denominated stablecoins on Uniswap, and used Liquity to leverage the remaining ETH as collateral for a loan of stablecoins.

We completed all of this in under an hour while eating lunch. We paid fees at each step in the process and we were happy to do so. We can see where all of the fees went and what financial services were being provided.

Account creation was free and instantaneous, thanks to cryptography. Transaction fees for Ethereum went to node operators who are securing our balances, processing transfers, and executing the code for the protocols mentioned. Liquidity providers on Hop and Uniswap received fees for allowing us to bridge the assets across blockchains and exchange the tokens in seconds with little slippage. Liquity charges no interest on the loan. There’s just a one-time borrowing fee provided to LQTY token stakers. (LQTY tokens are earned by those who provide LUSD to the Stability Pool, which helps maintain system solvency.)

I also didn’t need to take any notes throughout the process because all of the transactions were recorded on-chain. I just looked at Etherscan to confirm my memory of everything in the previous paragraphs.

The only reasonable criticism of this that I can imagine—from a systemic standpoint—was that our decisions were so risky that we should’ve been protected from ourselves. I don’t buy it.

Our decision to keep all of the funds denominated in dollars initially (rather than holding the original tokens and exchanging for dollars as needed) could ultimately result in a significant haircut on the project’s resources depending on future price action. We were comfortable with this because having more certainty around resources in the present is very valuable for decision-making.

Also, our loan risks liquidation. We felt that the liquidation price offered by Liquity was sufficiently low given historical price action and current market conditions. The software we’re building helps developers use Ethereum, so it seems sensible to have some alignment in the financial interests of the project.

Every decision involves some risk, including inaction. There was nothing amoral about any of the services provided by this open source software.

There’s a legitimate argument to be made that centralized services will be inherently more efficient, as trust could serve as some sort of shortcut. It seems to me that there’s actually a tremendous amount of rent extraction occurring in centralized systems and the perceived efficiencies are due to displaced costs. But, in any case, it’s an interesting point to discuss!

I also could imagine an objection: you don’t need a blockchain to solve these problems! This is fair, but who is you in this case? I personally can’t update Vanguard or Plaid’s APIs.

And of course I don’t mean to suggest these two anecdotes are representative of the vast array of interactions people can have with financial systems. Considerations around money, finance, and the economy touch all of our lives in infinite, complex ways.

I’m only illustrating here that the “political” debate needs to focus on the competitiveness of this new technology, while keeping in mind that it’s still nascent. There is no doubt that DeFi has a long way to go. But at the current pace of development, it really feels like it’ll be sufficiently robust and scalable for most use cases before the bureaucracy of large banks are able to ship a simple, open API. (This is assuming there’s even any motivation for them to accomplish this.)

The outcome here seems fairly simple to me. DeFi will continue to improve. As it begins to work better than alternatives, people will use it instead. I strongly hope that Americans will be encouraged to shape its development, despite some regulators’ and media outlets’ best efforts to force it abroad.

I understand that expecting critics to have ever executed a transaction on a blockchain is unrealistic. Though I also can’t help but wonder if they’ve ever been customers of a bank before.