Which stablecoin is the safest? (A deep analysis)

Could any stablecoin survive a recession?

One of the most relevant questions in the cryptocurrency market is about the safety of stablecoins.

In this article, I will not only show you which stablecoins are safe and unsafe, but I will also show you how to evaluate a stablecoin in practice.

On this image you are seeing the top 10 stablecoins on the market. In addition to these, I will talk about FRAX and Synthetix Dollar, because they are becoming popular.

It will be a article focused on stables linked to the dollar. Are you ready? So let’s get started:

First, we need to divide stablecoins into two major groups: Fiat-collateralized stablecoins, which supposedly have actual dollar reserves for each token issued, and algorithmic stablecoins, which maintain parity with the dollar using cryptocurrency reserves.

For the fiat-collateralized group, we must first ask whether dollar reserves actually exist. In theory, there will always be a company responsible for custodying dollars and issuing or burning tokens.

When this company receives a $1000 bank transfer, for example, it issues 1000 tokens to the user, where each token is worth $1, 

and keeps the actual dollars in custody. When the user wants his dollars back, he needs to return the 1000 tokens to the company,

which will burn them and return the real dollars to the user.

But why would anyone contact the company Tether Limited, for example, to issue USDT if he can buy USDT directly on an exchange? To avoid price swings.

The price of a stablecoin on exchanges can vary depending on supply and demand at the time, especially if an investor suddenly moves large amounts of money. But relax, even if the price fluctuates due to supply and demand, the tendency is for a backed stablecoin to maintain its peg over time.

 

As there is always the option for the investor to return the tokens to the company and receive their equivalent dollars. This is what gives the token its value.

Although the concept of fiat reserves is solid and brings security to the model, all risks are associated with the company that manages the stablecoin.

After all:

1. Who guarantees that the dollar reserves actually exist?
2. What financial assets are held as reserves?
3. Which are the main blockchains and exchanges where the token resides?

To answer each of these questions, we need to find:

• Audits and certifications regarding financial reserves.
• Secure assets equivalent to cash.
• Onchain data about the token.

As financial reserve is a lot of money, it is natural that the company tries to remunerate this capital in some way, which is why question number 2 is so important. It could be that the company lends money to third parties, invests in bonds or whatever, so we need to be aware of all the risks.

 

In the case of algorithmic stablecoins, the approach is different.

As cryptocurrencies are used as reserves, being automated and auditable on the blockchain, we can always know the current situation.

The custody of the reserves is managed by smart contracts, so the security is associated with the algorithm, 

as well as the blockchain used.

The main questions are:

1. Which cryptocurrencies are accepted as reserves for issuing the stablecoin?
2. How secure is the strategy to maintain the peg?
3. What happens if the peg is lost?

So now that we know what to look for, let’s start by evaluating fiat-backed stablecoins, because it is easier.

We first need to understand what kind of assets the companies hold as reserves and how risky each one is. The first three items are popularly called “cash-equivalents”, although corporate commercial paper does not have the same security as the others.

Money market funds are also considered very safe, but they are not the same thing as cash. I recommend reading this paper to be aware of the risks.

Following our list, security begins to fall exponentially, either by volatility or by asset quality. The list goes on with: Precious metals, Corporate Bonds, Funds, Stocks, Loans, Digital tokens, and so on. The risk of each one will depend on many factors, but the fact is that thinking of these last categories as being as stable and safe as cash is not appropriate.

Ok, so let’s find out the reality of each stablecoin by answering the 3 basic questions.

Tether Limited is currently audited by BDO Italia. In the past, it has been audited by 5 other firms.

Looking at the history, the first firm that audited Tether was charged by the SEC for “serial violations of the federal securities laws” and “improper professional conduct”. That is the first red alert.

BDO Global is one of the top 5 accounting firms in the world, but researching the history I found some scandals that the company was involved in.

 

On Tether, it has already been proven that there were irregularities in the past, and Tether paid a fine of $18 million for not having part of the reserves for the entire period of operation. While critical, that particular situation has put the market at ease, as many feared the problem could be much bigger.

Okay, we have already seen that there are some weak points regarding reserves.

On the second question, the distribution of capital is quite questionable, as about 18% of the reserve is held in the form of risky assets such as digital tokens, loans and corporate bonds, without specifying the particular companies and projects.

On question number 3, about 52% of the USDT supply in circulation resides on the Tron network, 44% on Ethereum. I have my reasons for finding Tron’s blockchain insecure, but I’ll leave that for another article.

Speaking of exchanges, the growing dominance of Tether on Bitfinex is another wake-up call.

In summary, we have that:

• The suitability of Tether Limited is questionable due to past events;
• The auditing of the reserves has a risk associated with the company BDO;
• The distribution of the token on blockchains and exchanges is not adequate;
• Around 18% of the reserves could become insolvent depending on market conditions.

In other words, I definitely do not consider it a good choice to use USDT as a store of value for the long term.

Well, it’s time to talk about USDC. USDC is managed by Centre, a consortium formed by Circle and Coinbase.

Like Tether, USDC’s balance sheet is audited by one of the largest accounting firms in the world. In this case, Grant Thornton. But unfortunately that firm is not scandal-free either.

In terms of prestige and credibility, Thornton appears slightly ahead of BDO.

One difference with Tether is that USDC’s audits are monthly and done by the same firm for over 4 years. USDC’s track record has no cases of misconduct or scandal so far.

On the onchain data, 90% of the supply resides on Ethereum, and the remaining 10% is spread across various other chains. That setup looks great.

The USDC balance on exchanges also looks relatively healthy, proportional to the volume of each player.

Regarding the transparency of reserves, 100% of the funds are held in US treasuries and cash, which seems to be a safe setup. 

This distribution has been much worse in the past, but Centre has recently tailored the balance sheet more diligently.

In terms of licensing, Coinbase is a heavily regulated publicly traded company in the US. 

Although the company is making a loss due to volume declines in the 2022 bear market, the company’s financial situation does not appear critical.

As a negative point is the fact that the company has introduced the term that, in case of bankruptcy, customers holding assets in the exchange will be considered unsecured creditors, i.e. they will be the last in the priority list in terms of restitution.

In summary, we have that:

• The auditing of the reserves has a risk associated with the company Grant Thornton;
• The distribution of the token on blockchains and exchanges is adequate;
• The distribution of the reserves in financial assets seems adequate.

It can be seen that the USDC stablecoin is much safer than the USDT, but it is still important to be cautious. Depending on the duration of the bear market, financial problems at Coinbase and Centre’s operation could eventually bring complications for USDC. Not to mention the basic regulatory risk that the entire stablecoin market faces. It is important to monitor closely. 

The upcoming quarterly reports will be important to also reveal the impact of USDC’s withdrawal as a trading pair on Binance.

Before talking about Binance USD, which is the next asset in terms of market value, it is worth talking about Pax Dollar, as the company Paxos is responsible for the custody of both Pax Dollar and BUSD.

So let’s get to it!

Paxos’ reserves has been audited monthly for over 4 years by the firm Withum, which is much less well known than Thornton or BDO. Perhaps because it is less popular, or on merit, I have not found any past scandals involving the company. Nor have I found any evidence of fraud involving the Paxos company.

The USDP token basically exists only on Ethereum, which is good.

About the balance sheet on exchanges, there is an exaggerated dominance of Binance.

About the distribution of reserves, the USDP balance sheet is composed 100% by US Treasury and cash, which is considered safe. Unlike Centre that changed the composition of its balance sheet over time, Paxos has maintained this policy since the creation of Pax Dollar.

In regulatory terms, the Paxos company is authorised by the New York State Department of Financial Services to offer USDP, Pax Gold and BUSD. 

The company has also secured other memorable licenses in other sectors, such as the SEC’s permission to Settle Equities on Blockchain.

In summary, we have that:

• The reserves are audited by Withum and regulated by the New York Department of Financial Services (DFS);
• The distribution of the token on blockchains is adequate;
• The distribution of the reserves in financial assets seems adequate.

It is worth highlighting its proximity to Binance, which despite being positive because Binance has validated and trusted Paxos’ service, brings a possible risk of contamination in case there is a problem with Binance, since it is the exchange that concentrates the most USDP. Another point is the fact that Paxos is a privately held company, which makes it difficult to know the current financial situation of the company.

Anyway, I would opt for Pax Dollar instead of BUSD, as BUSD inherits all the USDP risks, with no upside, and multiplies by the Binance risk. That’s two points of centralization.

Now that the Paxos service is understood, there is not much to talk about BUSD, because BUSD is the same as USDP plus Binance risk.

Only two backed stablecoins remain in our analysis: TrueUSD and Gemini Dollar.

TrueUSD’s reserves is audited by Armanino LLP, which is similar in size and credibility to Withum, which audits USDP. I also found no past frauds or scandals involving this firm or TrueUSD.

Armanino’s audit happens in real time and can be followed online. 

However, their report informs in a superficial way how TrueUSD’s balance sheet is distributed, even mentioning “highly liquid investments of sufficient credit quality” without informing exactly what they are about or what percentages are allocated.

TrueUSD is integrated with Chainlink to provide reserves guarantee in real time, so that, if at any time the reserves balance is lower than the token supply, everything is frozen. It is an interesting initiative, however it does not invalidate the lack of transparency regarding reserve investments.

TrueUSD claims to be regulated by the Nevada Department of Business, however I could not find an official document validating this claim. 

Regarding onchain data, TrueUSD’s distribution on blockchains has considerable exposure to the Tron network. Unfortunately I could not find token distribution by exchange, as it is less popular.

In summary:

• The reserves are audited by Armanino LLP;
• The distribution of the token on blockchains is questionable;
• The distribution of the reserves in financial assets lacks transparency.

Although it seems like a solid project in theory, the difficulty of validating information leaves the security of this stablecoin open in my opinion. The company could collaborate with more transparency and links to documentation that validates its statements.

Alright, let’s now review our latest fiat-backed stablecoin: Gemini Dollar.

This stablecoin is managed by Gemini, the company of the Winklevoss twins. GUSD’s backing is audited monthly by BPM LLP. It is a similar sized company to Withum and Armanino. I have not found any scandal associated with this accounting firm.

About Gemini, as a problem we have the recent CFTC lawsuit regarding bitcoin futures contracts offered in 2017. The consequences the exchange may suffer from this are still uncertain, although apparently unrelated to GUSD.

GUSD’s reserves are held in cash deposits, U.S. Treasury and Money Market Funds, without saying the proportions. 

The GUSD token exists only on Ethereum,

and Gemini is the dominant exchange, virtually isolated. Like Paxos, Gemini is regulated by the New York State Department of Financial Services.

In summary, we have that:

• The reserves are audited by BPM LLP and regulated by the DFS;
• The distribution of the token on blockchains is adequate, but there is little liquidity outside the Gemini exchange;
• The distribution of the reserves in financial assets seems adequate.

These characteristics put GUSD in a similar category to USDP and USDC.

Okay, we are done with the fiat-backed stablecoins. Before doing a ranking, we need to look at algorithmic stablecoins.

I will comment on these projects in a generic way, because a detailed explanation about each algorithm deserves separate articles, as I have already done about the DAI cryptocurrency and also about the DJED stablecoin that has not been released yet. I recommend reading those articles calmly later.

In summary, DAI was the first proof that it is possible to maintain parity in an asset without depending on the asset in question. Those who readed the articles on DAI and DJED realised that the biggest risk to a parity algorithm is when the market goes through a flash crash.

If the fall of the cryptocurrency serving as collateral is very large in a short period of time, the value of the reserves can become lower than the market value of the stablecoin. Depending on the situation, this can lead to a death spiral.

There is no magic, the only way to minimize the risk of insolvency is to have a large level of overcollateralization, i.e. far more reserves than the amount of stablecoin tokens issued. However this also reduces the efficiency of the process.

I particularly believe that the technical mechanism and the theoretical level of overcollateralization of the DAI and DJED projects are adequate. The way DAI and DJED deal with insolvency are similar: they use a second token to increase reserves. It is a clever concept, but not perfect.

I believe that the security of this model will improve when the cryptocurrency used as collateral, whatever it is, becomes sound money recognized worldwide. 

The day Ethereum, Bitcoin or any other crypto has massive adoption, being used as a store of value even for crisis scenarios, algorithmic stablecoins will become very secure. In the current scenario, I will make brief considerations about each stablecoin.

DAI, besides having a suitable mechanism, is the most time-tested. As a negative point is the increase in the amount of USDC in its reserves, which ended up inheriting centralization and custody risks of fiat-backed stables. Besides, the project has already announced that DAI will abandon the dollar peg in the future, changing its mechanism, which should take about 3 years to happen. The project is also allocating funds in US Treasuries, questionable decisions for those who started with a principle of decentralization and alternative to the traditional financial system. In other words, for the long term, DAI brings a lot of uncertainties.

DJED looks very interesting, but I wouldn’t use it as a reserve without waiting to be proven by time and adverse scenarios, after all the project hasn’t been launched yet. The same goes for very new stables.

Following our list, we have FRAX. FRAX is a stablecoin that works with a hybrid system, using both stablecoin reserves and a price stabilization based on another FXS token. This protocol deserves a separate article and promises to evolve over time, however as currently the reliance on USDC is still large (close to 90%), it might be more prudent to expose yourself directly to USDC.

Next on the list is USDD, which is an arm of the TRON project. That alone should already be a danger sign, but it gets worse. The concept of USDD was born along the same lines as LUNA UST. However Justin Sun has been trying to add real reserves that do not rely solely on TRX to USDD, as LUNA was trying to do and broke down before completion. With only a few months of existence, USDD has already significantly lost the peg at least twice. Staying away is the best thing you can do.

The next token is also questionable. Neutrino USD, built on the WAVES blockchain and using WAVES as collateral has an unrobust parity mechanism, which is reflected in a very bad stabilization. Being constantly far from the value of 1 USD, the project has flirted with death several times.

In the list of the top 10 stablecoins, only FEI USD was not mentioned. FEI deserves a separate article to explain its dynamics, which uses Ethereum as collateral, but instead of relying on another token to maintain parity in cases of insolvency like DAI and DJED, FEI works with punishments for selling the token. 

The further away from parity, the more FEI USD are burned from the portfolio of those who decide to sell. It is a differentiated strategy that also needs time to prove itself. For now, things have not been very good for the project. The interest and volume for this stablecoin has reduced, and after more than 1 year of operation, FEI USD has been trading around $0.98. 

As promised, I will still briefly mention sUSD, as it has gained popularity. sUSD is part of the Synthetix project, which issues many types of synthetic tokens. It has the largest overcollateralization on the market, however these collateral are entirely based on the SNX token, which makes the concept more risky. I particularly think that a collateral needs to have value in itself, separately and independently from the issued stablecoin. If the value of the collateral is directly associated with the issued stablecoin, a dangerous cycle is closed, similar to the LUNA UST case. Although not exactly the same, the similarity makes me think it is more prudent to wait a long time until this project proves itself.

Some years ago, after having studied the UST model, I decided not to invest in this kind of stable. I don’t regret it.

Well, now that we have covered all the main stablecoins in today’s market, I will make a summary:

I am a big believer in the concept of algorithmic stablecoins and I see a great future, but there is still a lack of reliable players. At the moment, no algorithmic stablecoin available on the market meets my criteria for long term hodl. If you need it in the short term, I would recommend only DAI and FRAX.

On fiat-collateralized stablecoins, I would go as follows: 

• USDP, USDC and GUSD are the safest. If you need to keep money in dollar-linked crypto, diversify between them. Don’t put everything in the same stable, and understand that none of them deserve your blind trust. If you plan to hodl, keep up with news, balance sheets and updates on these projects on a monthly basis. Any sign of danger, migrate. 
• Taking it this way, you can carry your money in stablecoins for longer periods with low risk, TrueUSD would be the next candidate on the list. From here on down, forget hodl.
• If you use Binance to do crypto buying and selling, I don’t see a problem with using BUSD in there. But just as it’s not wise to keep your tokens in exchanges, it’s not wise to keep in BUSD, especially knowing you can eliminate a layer of risk with USDP instead of BUSD.
• USDT can be useful for immediate liquidity and practicality as it is the most traded currency in the market today, but don’t consider using USDT for more than a few minutes or hours. Make whatever trade you need, transfer, swap, whatever, and get out of risk immediately. No medium term strategy should involve USDT.