Screen Shot 2015-11-18 at 4.55.47 PM

Compare the banking landscape today with that of 10 years ago, and it’s hard to miss the changes. Lenders typically hold more capital and are much less reliant on unstable wholesale funding than they were before the crisis. Regulators are generally warier of mounting risks in the financial system — whether these come from consumer debt or derivatives exchanged over the counter.

Yet the right question to ask is not what’s changed, but if these regulatory transformations have been sufficient. In at least three areas — the right level of bank capital, the use of risk weights, and structural reforms — many economists fear the financial system remains exceedingly vulnerable to shocks.

The largest gap between academics and practitioners is probably on the level of capital. As Sir John Vickers, a professor of economics at Oxford who presided over Britain’s Independent Banking Commission (IBC), noted in a recent speech, regulators are now accepting a level of leverage which is still around 25 or 30 times a bank’s core capital. Many outside economists believe a bank should only hold assets worth six to 10 times their key funds, if not less. “So one group or the other, if not both, would appear to be wrong by a large margin, on a policy question of deep importance,” Sir John noted.

Of course, most advocates of higher capital requirements don’t believe banks should get to the new ratio overnight. Lenders would do so by shedding assets rather than raising new capital, with dramatic impacts on the economy. But the question matters since regulators appear increasingly comfortable with the ambition of the existing rules and don’t want to go further. Vickers refers to the case of the Bank of England, which last year decided not to ask Britain’s largest lenders to raise significantly more capital over time, as it felt satisfied, among other things, with their plans for orderly resolution. It is hard to escape the feeling that some central bankers have become complacent over the level of risk they are willing to tolerate.

A related matter is our assessment of bank risk. At a conference held last week by the Centre for Economic Policy Research (CEPR), Tamim Bayoumi of the International Monetary Fund showed how the Basel Committee’s decision in 1996 to allow banks to use internal models for their valuation of risks radically changed the behavior of lenders, especially in Europe. Before 1996, banks with a higher ratio of risk-weighted assets to capital were also those with a higher leverage ratio. After that year, the relation between the two broke. Most likely, banks learned to game the system and pile up on assets simply by tweaking their risk models.

There is no guarantee that simpler leverage ratios can avoid a new financial crisis. After all, banks may simply hold on to their riskier assets and shed less remunerative but safer loans. The Basel Committee is introducing a simple leverage ratio as a backstop measure to internal risk models. Yet that doesn’t justify the prominence regulators continue to give to an approach which has shown manifest weaknesses during the crisis. Risk-weights, as the financial crisis made plain, can be pro-cyclical; risk falls when the level of an asset goes up and vice versa. The result is that not only are we asking banks to hold too little capital, but we are also underestimating how problematic their exposures really are.

The fear, therefore, is that sooner rather than later, governments may again be called upon to rescue a troubled bank. And here lies the third dangerous similarity with the pre-crisis world: Many Western countries have proven unable to separate investment banking from commercial lending. As a result, even if governments only wanted to keep the latter going, in many cases they will be forced to rescue everything, as it is impossible to split the two.

The worst offender is undoubtedly the European Union. In 2012, Erkki Liikanen, the Governor of the Bank of Finland, produced a reportrecommending, among other things, the separation of trading activity within universal banks. Five years on, the EU has failed to follow up on his suggestion in any meaningful manner, leaving mega-banks such as BNP Paribas and Deutsche Bank unchecked.

The U.K. and the U.S. have undoubtedly moved further in this respect. Britain is pressing ahead with the recommendations issued in the IBC’s report, including building a ring-fence between investment and commercial banks when they are in the same institution. In the U.S., the Dodd-Frank act has led to the imposition of the so-called Volcker rule, prohibiting banks from engaging in proprietary trading under certain circumstances. Yet, at least in the U.S., the administration is now considering a dilution of these measures, which could turn the clock closer to the pre-crisis years.

When judging the shape of financial regulation after a crisis, we often hear the industry view that the new requirements have been overly burdensome. There is however another, equally plausible take: that we have not gone nearly far enough in shoring up the banking system. Trade-offs are always difficult to assess, but if this more conservative assessment is right, it is a terrifying prospect.