Just because some government action or regulation would be bad, doesn't mean that any would

With regard to Arnold Kling's post, "Mark Thoma's Question", there's a few things I'd like to note:

"Moreover, to the extent that there are problems with executives looting corporations, I doubt that government is the solution. If you want to see real looting, watch what happens when Progressive Corporatism kicks into gear."

Yes, there is a lot of government action that would be bad, but that doesn't mean that all government action would be bad, or that all government institutions or regulations are bad. There is much that has been shown to be enormously good. So just don't do the bad government actions, but don't not do any, even the enormously efficient and good. Any well trained economist knows that there are market problems which can cause enormous inefficiency without a government role, such as externalities (especially the gargantuan positional/context/prestige externalities, the pink elephant of economics), asymmetric information, inability to patent, inability to perfectly price discriminate or come anywhere close, gigantic economies of scale and at the same time the gross inefficiencies that come with monopoly, and many more well established and accepted in economics, and in any college economics texts.

Note also that government has improved greatly over time with experience, academic study, and other learning. Look at how much more professional and objective the civil service had become over the last 100 years, unfortunately until the conservative Republicans predominantly took power 28 years ago.

"My idea of what to do with insolvent banks isn't to inject capital. My idea is to close the doors and tell the executives, "Don't let the door hit you on the way out."

If there are banks that are borderline because their securities are not trading, then put them on a watch list, quarantined from other banks, until their situation clarifies. My hunch is that the healthy portion of the banking system is sufficient for the economy, but we have to get rid of the zombie banks first."

Why would the zombie banks be such a problem in preventing the healthy portion of the banking system from providing adequate liquidity, if that healthy portion is, as you have a hunch, sufficient? Clearly from the various financial measures reported, there is, and especially was before the recent government actions, a very serious liquidity problem.

In any case, a hunch is not strong enough to risk a severe recession or depression for, not given the relatively small expected cost of acting in the way Krugman and Eichengreen recommend. If you have a hunch that you will not have a fire, that's not a strong enough reason to risk not buying fire insurance. If there's even a 1% chance of having your house burn down, that makes it worth buying insurance, even though 99% of the time you'd be able to say, "Ha Ha, see my hunch was right."

As far as letting the bad companies go bankrupt, I basically agree, but in a fast track way – i.e., the government takes them over, gives the shareholders nothing, if the firm is really worth nothing, gives the bondholders only what their bonds are really worth, no more, and tells the executives, "Don't let the door hit you on the way out.". But then the government keeps at least most of these companies running, so as not to risk a "house fire" over your hunch. It injects capital, cleans up the balance sheets, and then re-sells the newly healthy companies back to the private sector.

If instead you let these companies go bankrupt the conventional way, it could take years for them to complete the standard legal bankruptcy process, and for other companies to take over their earlier liquidity provision. Waiting for the long run means potentially allowing a severe credit crunch to start a demand-decrease chain reaction that leads to an acute recession or depression.