You Can Always Borrow More. Moving It Costs Money You Do Not Have.

Your company holds a treasury. Some months ago you borrowed dollars against it, because selling would have been a taxable event and a signal you did not want to send, and a loan was neither. You picked a venue, you accepted a rate, and the position has been sitting there ever since.

Somewhere in those months a better rate existed. You noticed. You have thought about moving more than once, you know roughly what it would involve, and you have not done it.

That was not carelessness. It is the shape of the thing you are holding.

A borrowed position has exactly two adjustments available to you, and they cost wildly different amounts.

Making it bigger takes one action. Nothing has to leave, nothing has to be found, nothing has to be timed. The collateral is already posted, so you borrow more against it and you are done.

Moving it takes the debt itself. To leave a lending market you have to pay it back, and to pay it back you have to be holding the dollars you borrowed. If you were holding those dollars you would not have taken the loan. That is the trap, and it gets worse with size rather than better: the bigger the position, the more absurd the first step becomes.

There is a way around it, and it is the reason this is worth measuring rather than asserting. You can borrow the payoff amount for the length of a single transaction, use it to close the old loan, open the new one, and hand it back before the transaction ends. Nobody ever holds the cash. It works, it is not exotic, and it is how every serious version of this is done.

So the question is not whether moving a loan is possible. It is how many people actually get to do it.

We read every new loan and every repayment on Aave and Morpho — two of the largest lending venues on Ethereum — across ninety days, ending at a single pinned point on the chain. Not a sample. Sixty-eight ranges, every one of them answered in full, none dropped and none scored as empty because it failed.

In that window, 14,723 separate wallets borrowed money.

Then we looked for the move: a repayment at one venue and a new loan at the other, inside one transaction, which is the only way to do it without ever holding the cash.

Seventeen wallets went from Aave to Morpho. Sixteen went from Morpho to Aave. Eight of those appear on both lines, so the total number of separate wallets that moved a loan between the two, in ninety days, is twenty-five.

Not twenty-five thousand. Twenty-five.

Maybe nobody wanted to move. That is worth taking seriously, and the same data answers it.

Over those ninety days there were 76,343 new loans opened on Aave against 58,731 repayments, and 32,736 against 25,798 on Morpho. Nobody is sitting still. These positions get touched constantly — they are just only ever getting touched in one direction. About thirteen new loans for every ten paid down, at both venues, arrived at independently.

Two venues that share no code, no pricing machinery and, mostly, no users, landing on the same ratio is not a quirk of one place. It is what the shape of the choice does to everybody standing in front of it.

One honest limit. A large share of borrowers at both venues made no repayment at all in the window — and on Aave that share depends heavily on how busy the wallet is, falling from 54.3% of wallets that borrowed once to 16.8% of those that borrowed eleven times or more. On Morpho it barely moves, 50.6% down to 41.1%. So "never paid anything back" is partly a statement about people who borrowed once and were still holding the loan on the last day we looked, which is a normal thing to be doing. We are not going to build an argument on it. The twenty-five is the number that survives every way of slicing it, because it is a count of wallets and nothing else.

It is not that this problem is unknown. Euler shipped a guided flow for exactly it, and put it in their own words:

That is a real feature and it does the job. It also moves debt between Euler's own vaults, which is the shape every version of this has: the tool belongs to the destination. It is built by whoever wants your loan to arrive, which means it exists exactly where somebody had a commercial reason to build it, and nowhere else.

You can see what that produces. Moving inside a single venue is the easy case, the case the venue itself has an interest in making work, and even there it is 977 of Aave's 10,978 borrowers and 42 of Morpho's 3,745.

You do not need our numbers for this part. Take the borrowing on your own balance sheet, the real figure, and ask what you would have to hold in cash, at one moment, to move it somewhere better.

All of it. Not a fee, not a spread, not a few pennies of friction — the entire principal. That is the price of changing your mind about where your loan lives, and unless somebody has written you the machinery to borrow it for the length of one transaction, that is the price you are quoted.

Now put that next to the price of the other adjustment, the one that makes the position larger: nothing.

This is the part that matters, and it has nothing to do with rates. When one direction is free and the other one is priced at the whole loan, you do not end up with the position you chose. You end up with the position that the cost of moving built for you. You size up, because sizing up is the only lever that ever works when you want to improve the economics. You hold through a rate you would have left if leaving were cheap. Every adjustment you make is the risk-adding one, because it is the only one available, and after a few years of that you are looking at a book you would never have designed on purpose.

Nobody set out to be over-levered. They set out to make the one free adjustment, repeatedly.

In the United States, 103 distinct phrasings of one question — can I refinance without paying for it — get searched 17,250 times a month. The head terms clear eleven dollars a click.

We checked twelve ways of asking the same question about an onchain loan. Eleven came back with no row at all and the twelfth came back at zero. Not a low number. No number.

Those searchers are shopping for mortgages, and we are not going to pretend they are your customers. That is not the point. The point is that "how do I move this loan somewhere better without paying for the privilege" is one of the most ordinary questions in consumer finance, and onchain it has not been asked often enough to register as a question at all — while twenty-five wallets quietly do it.

If moving costs nothing, the decision changes shape. You stop sizing up to justify a trip, because there is no trip. You set the condition once — the rate that would make you leave — and the loan goes when the condition is true, whether or not you are looking at it that day, and without you ever holding the money you owe.

The position you end up with then is the one you actually chose.