The proposed $700 billion bank bailout is stalled in Washington, though as of early Friday negotiations (and finger-pointing) were still ongoing. Perhaps that's for the best. At this point, I'm on the fence about whether we'll be better off with the bailout or without it.

It could help. It should help. I want to think that we saw the bottom in stocks last week.

I'd be more encouraged if the recent bailouts of Fannie Mae (FNM), Freddie Mac (FRE) and American International Group (AIG) had been more successful. But they weren't. They ended up making matters worse, requiring now an even bigger bailout.

So what happens if this one fails? Do we do another one, even bigger yet? Already in the interim the government had to cobble together a rescue of Washington Mutual (WM), which late Thursday sold to J.P. Morgan Chase (JPM) for $1.9 billion.

That's the way it is with government, and why government is fundamentally different than private enterprise. When private enterprise fails, people don t repeat the mistake. When government fails, people do repeat the mistake -- only bigger.

Yes, the situation we are talking about here is a strange blend of the private and the governmental. In one sense we wouldn't be talking about a bailout at all if the private sector hadn't screwed up. But I never said the private sector was perfect. I just said it doesn't often make the same mistake twice. That's government's job.

And when the private sector knows that a bailout is possible, then the disincentive to make and repeat mistakes is lessened. So more mistakes are made, only bigger ones. In fact I could make a pretty good case that the current crisis is precisely the result of a deadly blend of private-sector greed and public-sector foolishness, reinforcing each other in the worst possible way.

But that doesn't mean that eventually we won't figure it out and get it right. What are the chances that we're going to do that with the massive $700 billion bailout now being debated on Capitol Hill?

As a matter of first principles, the underlying idea of the proposed bailout is a smart one. There are impaired assets on the balance sheets of banks and brokers. There's no ready private market for those assets, except at ruinous prices. But that doesn't mean the assets are really worthless -- it just means no one wants them, because everyone already has too many of them.

So it's a perfect time for government to step in and help. Well, strictly speaking there's no such thing as "perfect" when it comes to government interference in the economy, but if ever there was a good time to do it right, this could be it. Here's why.

If the federal government were an investor, it would be the most efficient one in the world. It has the lowest cost of funding to buy investments, because it can raise money by selling Treasury bonds at low interest rates, while banks and brokers have to pay high rates to get their hands on borrowed money. And because of its vast size, and the diversity of its income streams and asset holdings, the government is in the best position to take on risk.

That means that the government ought to be the highest bidder in the market for risky investments because those investments are worth more, and present less overall risk, to the government as compared to any other possible investor.

So if government has any role at all in bailouts, this is it. Not just to print money and give it to distressed banks and brokers, either directly, or through tax breaks, or anything else. Instead, it should simply step up and buy assets that are a glut on the market in order to help the U.S. banking system get back to the business of making loans to people and to businesses.

As many commentators have said this week, this way of thinking about it means that the government is very likely to make money on this by the time all is said and done.

But there's a paradox here. For the government to make the most possible money, it has to buy the distressed assets from the banks at the lowest possible price. If it does that, then it's hardly a bailout, is it? The whole point is for the government to pay more than the banks could get if they went to the open market. But then that cuts into the government's potential profits.

Actually, markets are very good at solving problems like that. Every time a trade occurs between buyer and seller, a bargain was struck that somehow resolves the conflicting needs of both. (If it didn't, the trade wouldn't take place.)

The problem is that the Treasury has no clue about how it's going to organize the market for transferring distressed assets from banks to the government.

They keep talking about a "reverse auction." That just means that all the sellers compete to make the price lower and lower, and the buyer eventually buys at whatever price the sellers determine through that competition. That process would seem to be aimed at producing the lowest possible price, which means it will minimize the benefit of the bailout for the banks.

Which means it won't achieve its stated goals: to help the banks.

There's a lot of noise to the effect that if we help the banks, we have to punish them at the same time. Make sure their executives don't get paid. Take an equity stake, as insurance that the taxpayers won't lose on this deal.

But you can't save someone and punish him at the same time. If we're going to do a bailout, let's do a bailout, not a lynching.

We really need to stop and think these things through. Nothing good ever comes from haste. Haste implies lack of knowledge. Any success in any field -- other than buying lottery tickets -- comes from knowledge. Federal Reserve Chief Ben Bernanke and Treasury Secretary Hank Paulson have none. Congress has none.

I don t say that to criticize them, only to caution them to be duly modest here. Admit what you don't know. Start slow. Learn by doing.

The idea of spraying $700 billion at the market with a firehose, and having no idea how to direct it to the best use to solve the credit crisis, is sheer idiocy and will end in tears. And the idea of rushing this legislation through, loaded up with a lot of punitive anti-Wall Street provisions that nobody has really thought through, just makes no sense at all.

Bottom line: This bailout could work. It should work. And it may actually work. But until we know what it entails, we can't be anything close to sure about it. So the most bullish thing now would be for Congress to approve the bailout, but be sure that Paulson and Bernanke honor the need to start with baby steps. Congress, too. It shouldn't insist on a lot o punitive features until it sees who really needs to get punished, and for what -- if at all.

If they do that, we'll have this economy back up on its feet in no time. But if they blunder into this without a plan, they will absolutely for sure make things worse. I'd rather see no plan at all than a bad one conceived in haste. I have a feeling that the stock market will see things the same way.