Much of the historical debate surrounding Ethereum ETHUSD There has been speculation about whether gas fees are too high. This year, thanks in part to rollups, fees have become extremely low – except during the crash of August 4th and 5th, when they were disproportionately high. So what is the reason?

The surprising answer is that everything worked as planned. Periodic spikes are the behavior of a wholesale market during peak demand. Those who call these developments problematic have the wrong vision for the future of cryptocurrencies.

The modular scaling philosophy is based on the idea that blockchains are not networks that process activity, but rather suppliers of a scarce asset called secure block storage. Like any other scarce asset (like land, oil, or electricity), secure block storage is continuously auctioned off to the highest bidder, and the winner is the one who needs it most. This will likely be either a whale or a wholesale buyer, such as a Layer 2 buyer.

Related: Forget Ethereum ETFs – Here’s What You Can Do Instead

Monolithic chains have more of a network philosophy: they want to serve everyone on an equal footing. This is unsustainable because blockchains are not just networks, like the internet is a network. If a video stream or an email is delayed, it’s no big deal. But if an important financial transaction is delayed, it could be catastrophic. Blockchains are predictable and secure transmission and settlement layers. In any other context where such a service is offered, both the service and the cost are tiered.

Imagine if the post office charged the same for a second-class postcard as it does for an express package. That couldn’t possibly work because everyone would choose the latter, even for non-priority packages. If supply is always less than demand, you have to keep adding capacity to prevent rates from rising ever higher.

But on a modular chain, it doesn’t matter that much. Individual users who still trade on L1 are there because they want maximum security. This only makes sense if they are doing large transactions. A trading firm doing a $5 million swap doesn’t care if the gas price goes up to 500 Gwei. It has little impact on their bottom line.

The other players on such an L1 are wholesale providers such as rollups. These players also don’t care so much about short-term spikes in base fees. They have a healthy profit margin when activity is normal and fees are low. This creates a buffer for them to operate at a loss during spikes.

This is how any other market for scarce goods works. Airlines and gas stations are major buyers of oil products. When oil prices are low, they have a higher profit margin. When oil prices rise, they absorb most of it and may operate at a loss. If prices remain high, they slowly drive up their own prices. The important thing is that private customers (airline passengers, car drivers, etc.) are one level further away.

Related: Expect Bitcoin ETF options to launch before 2025

The caveat to all this in the crypto world is that during the August 4-5 crash, rollup gas fees also skyrocketed. If my theory was correct, that wouldn’t have happened. But that’s because the rollup fee market is still very immature. Over time, competition will force L2s to compete on price. Not just on absolute levels, but also in terms of volatility. Some will even advertise themselves with this promise. When that happens, sequencers will set aside some of their profits in the good times to cushion shocks in the bad times.

I’m confident that most gas fees will eventually be absorbed by end users, especially retail users. We’re already seeing this with some things like USD Coin (USDC) transactions on Base. Part of the appeal of launching your own L2 is the ability to treat gas costs/sequencing as a lead offer. The same will happen with dApps (One example is UniswapX, which takes a portion of gas fees due to MEV) and even wallet providers. This UX evolution could still happen on a monolithic chain, but is more stable on a modular one.

Omid Malekan is a guest columnist for Cointelegraph, an adjunct professor at Columbia Business School, and author of “Re-Architecting Trust: The Curse of History and the Crypto Cure for Money, Markets, and Platforms.”

This article is for general information purposes only and is not intended to be, and should not be construed as, legal or investment advice. The views, thoughts and opinions expressed here are solely those of the author and do not necessarily reflect the views and opinions of Cointelegraph.