Academic

They are pretty worried in Europe that the answer might be yes, and this has been a problem with getting Basel III done.  This sort of question, which involves always unpredictable macroeconomic forecasting and on which the fate of the party in power 100% depends, isn't really amenable to academic study.  Nonetheless, that's what those research departments at central banks are for, and so some of the Basel Committee's people, and the IMF, did some studies. 

It's hard to believe, but the studies showed that the thing Basel and the US wants to do - increase capital adequacy requirements - won't be bad for the economy.  Instead, they'll be long term good for the economy!  Hurrah, but the interesting thing is that even these guys admitted that there would be short term pain (though they think it will be miniscule):

The Basel Committee's assessment of the long-term economic impact finds that there are clear net long term economic benefits from increasing the minimum capital and liquidity requirements from their current levels in order to raise the safety and soundness of the global banking system. The benefits of higher capital and liquidity requirements accrue from reducing the probability of financial crisis and the output losses associated with such crises. The benefits substantially exceed the potential output costs for a range of higher capital and liquidity requirements.

The FSB-BCBS MAG assessment of the macroeconomic transition costs, prepared in close collaboration with the International Monetary Fund, concludes that the transition to stronger capital and liquidity standards is likely to have a modest impact on aggregate output. If higher requirements are phased in over four years, the group estimates that each one percentage point increase in bank's actual ratio of tangible common equity to risk-weighted assets will lead to a decline in the level of GDP relative to its baseline path by about 0.20% after implementation is completed.  In terms of growth rates, this means that the annual growth rate would be reduced by an average of 0.04 percentage points over a four and a half year period, with a range of results around these point estimates.