Squeezing those excesses out of the market is a necessary move to detoxify the rest of the economy. Most importantly, it will force the banks to reorganize their business.

This can't be called an austerity program, because we're talking about a long, slow process without a crash. Nor, certainly, is it quantitative easing. It is a deleveraging campaign.

Actually, it's only recently that people can read news about funds in China collapsing, debtors defaulting and companies failing.

Getting those results has taken a GDP growth rate of less than 8 percent from the first quarter of 2012 to the fourth quarter of 2013.

If China can, in two more years of still lower growth (GDP expansion of 7.5 percent or less), it can reasonably hope to phase out a significant part of its old-economy industries.

And if, as some business commentators propose, the country is to eliminate up to half of its industrial base to make way for business based on the mobile Internet, it will have to keep the credit line really tight for local development projects.

The central government has declared that it will build a municipal bond market. But the period from passing laws and regulations to witnessing a functioning market could be at least two years.

It will take at least as much time to establish a credit rating system just at the city government level. To extend the system to all local governments, from the provincial level down to the township level, will require a much larger and longer effort.

So, with perhaps only a few exceptions, local governments will face a long period of tight credit. Many mayors and other officials will find it hard to achieve their development goals before their terms expire. Their most ambitious projects will be starved of funding.

People traveling to Haikou, Hainan province, and Beihai, in the Guangxi Zhuang autonomous region, can still see some "rotten-tail" (unfinished) buildings from property booms in the 1990s.

A boom cut short can leave nearly permanent scars on a city. There will probably be more cities in just that situation a few years from now.

Just as the central government has never helped Haikou and Beihai rebuild, it is all up to local governments to redirect their development path.

Regardless of how strong China's central government is, the things that it can do are limited. It can grant free trade zone status to a city such as Shanghai. But it doesn't have the resources to support a few hundred cities all at once, especially when they are all making the same mistake.

There's a rumor that a new round of stimulus is coming soon, since indicators have shown that the economy's first-quarter performance might have been close to the worst in the past couple of years. The premier heard a lot of complaints in his latest fact-finding trip to Northeast China.

But even if the central government wants to give a boost to growth, it will have to be a different kind of stimulus package from the one that former Premier Wen Jiabao signed in late 2008, shortly after the Wall Street meltdown. Some key points:

It can't be as excessive as the 2008 stimulus package, to which the central government alone contributed 4 trillion yuan (close to $650 billion), and local governments contributed even more (mostly on credit, of course). A large stimulus plan would easily discredit the government's reform commitment in the eye of economists.

It can't be entirely directed to large State-owned enterprises, which will use the money to build unwieldy industrial monopolies that will only have to be broken down later. It cannot be allowed to stunt the growth of other types of businesses.

It can't be a purely monetary package, and it must integrate policy changes. A stimulus that works better for service industries, private-sector companies and small and medium-sized enterprises will contain more policy incentives, not just credit incentives.

It can't directly support local governments' large ongoing projects. To keep financing them, local governments will have to learn to be responsible bond issuers and communicate with investors, instead of the premier, in the market. Local governments grappling with too many ongoing or unfinished projects will have to learn how to privatize them through lawful channels.

It can't enable real estate development (meaning expensive housing projects) to dominate the economy again.

It can't compromise the health of the financial industry or lead to a surge in nonperforming loans. Nor should it trigger a quick rise in inflation.

Having set all these conditions, whatever stimulus might be introduced must be a strictly central government program, targeting select industries and locations.

It must facilitate, rather than to run counter to, the goals of reform.

In the meantime, most or all companies that still operate in the old industrial economy and pursue obsolete development strategies will face a long, painful path to reform.