FIXING the banking system to prevent another crisis on the scale of 2007-08 is a fiddly and time-consuming task. It is not the kind of thing that generates tabloid headlines and public approbation. But another important stage in the process was reached on November 9th when the Financial Stability Board (FSB), a global regulators’ forum, issued new guidelines on bank balance-sheets.

The underlying problem is as ancient as banking itself: banks lend out more money than they have capital to absorb losses. If their loans go sour (or if the banks’ own creditors, including depositors, lose confidence), institutions can rapidly go bust. And then, because of the importance of the banking system to the economy, governments feel obliged to ride to the rescue at potentially vast cost.

To avoid this danger, the FSB wants private investors to bear the cost of bank failure. As well as their equity capital, banks should issue a new type of debt with terms that make it explicit that the lenders will be among the first to take a hit if the bank gets into trouble. This has two advantages. First, severe bank losses will be absorbed by these bondholders and not by depositors. This reduces the risk of a public panic and thus the need for a government rescue.

Second, the threat of a potential loss will encourage bondholders to keep a close eye on bank management, and to intervene if they feel executives are taking too much risk.

Asking banks to raise a lot of new debt immediately is not ideal. One of two things might happen.

The banks could flood the market with new bonds, causing the cost of borrowing to rise. Or the banks could improve their capital ratios by lending less, cutting off funds to business and hurting the economy. So the reform has to be phased in.

All these changes sound eminently sensible. But this is finance, so the regulator’s proposals inevitably come cloaked in impenetrable jargon dotted with obscure acronyms. The relevant measure of capital (equity plus the at-risk debt) is called “total loss-absorbing capacity” or TLAC. The ratio will only apply to the most important institutions, dubbed global systemically important banks (G-SIB). It is hard to see the public marching behind Mark Carney, who heads both the FSB and the Bank of England, shouting, “What do we want? Higher TLAC for G-SIBs. When do we want it?

Phased in between 2019 and 2022, except for emerging-market banks which will have till 2025 to 2028 to comply.”

Those familiar with the minutiae may complain that the FSB has not been as demanding as it might have been. It is asking banks to have TLAC of 16% of their risk-weighted assets by 2019, rising to 18% by 2022. This is at the lower end of the expected range. The FSB estimates that banks will have to raise as much as €1.1 trillion ($1.2 trillion) in new TLAC-compatible debt.

But not all of this will be additional borrowing: banks can simply issue TLAC bonds to replace existing debt as the latter matures.

That helps explain why the estimated costs of the reform are so low. Doubtless, investors will demand a higher yield on bonds that will be counted as TLAC. This higher cost is likely to be passed on to bank customers in the form of higher lending rates, which may crimp economic growth. But the FSB estimates that the effect, in both cases, will be a small fraction of a percentage point—far less than the costs of another bank blow-up.

None of this will prevent a future banking crisis. Instead, the aim is to limit the damage such a crisis can create by making clear where the losses will fall—something that was not obvious in 2008. The biggest economic impact occurs when depositors, having assumed their money was safe, take fright; that is when you see queues round the block. Back in the 1930s America introduced deposit insurance to reduce this problem. The new reforms give depositors an even more secure status.

But bank losses have to be borne by somebody. In that sense, bank runs are rational: if you think a bank will run out of cash or capital, it makes sense to withdraw deposits, or to dump investments in the bank, before the rest of the herd. That will be true of TLAC bonds, too.

Alas, you can’t get rid of risk altogether. Because TLAC bonds will take the hit during the crisis, they will provide early warning of trouble. But corporate-bond markets are pretty illiquid, and even liquid government bonds have seen sudden lurches in price. So it is quite possible that TLAC bonds could give false indications of a looming crisis; smoke without fire, in other words. Even sensible reforms have their flaws.