I'm not so sure I agree with the "do no harm" theory. If somebody gets control of a government and starts doing bad things, at what point is intervention required? Perhaps an example is to compare Hitler and Stalin. Both were terrible dictators and killed millions of people, but in Hitler's case the world cooperated to shut him down, while in Stalin's case the world let it ride. Is that latter model the one we wish to follow? Just let a dictator murder people, because it is too hard for us to agree about the right way to do something about it?

I suppose one can say that yes, the best strategy is to allow each country to run its own internal affairs however it likes, and not worry about what goes on inside. But then one must not listen to whining about the civilian populations of Darfur or North Korea or where-ever--just let them rot.

Some points to further your position.

Contrary to war apologia spread by Clinton, Wesley Clark et al, Milošević was actually temporarily strenghtened during the war. This was natural, it happened so before and after elsewhere, but the advocates of air war never learn: people will pull together when bombs are falling. So Milo could actually step up his hunt for oppositionaries, some of whom had to go in hiding, and even lost popularity for doing so.

Furthermore, the West did have a role in Milošević's overthrow, but not via the war: it was the concerted effort to get the opposition to organise itself and unite (this went along two main routes, financial and material support in the form of sister city help to local governments under opposition control, and organising conferences -- the most important here in Budapest -- to get the different factions to cooperate).

And the long-term consequences for Serbia proper (they never consider the long-term consequences) were even worse than you describe. That chief propagandist of 'liberal' interventionism, Thomas "Let war give a chance!" Friedman, infamously wrote: "Every week you ravage Kosovo is another decade we will set your country back by pulverizing you. You want 1950? We can do 1950. You want 1389? We can do 1389 too." 'They' did 1950. Sane people realise that the country doesn't consist of Milošević, and realise that throwing a country back decades means poverty and resentment lasting for decades. Especially if once pro-Western forces get into government, and start reforms (including neoliberal "reforms"), yet the West won't send the promised financial aid, resentment will turn into hate and into support of demagogues of various dangerous kinds. This is exactly what happened. The air war killed any possibility of a genuine development towards Western-style democracy.

_{*Lunatic*, n.
One whose delusions are out of fashion.}

The bigger mess 'theory' is a rather simple rule of thumb, I'll grant. It's attractive because it can be tested, to a degree.

What you want to know, of course, is what course of action would have been best, which would require plotting various scenarios, which is a lot of work and largely untestable. But the bigger mess rule ignores the hypotheticals that a situation is on the brink of spiralling beyond control and, on the opposite, that a better solution is clearly available.

All ethnic Serbs are not yet out of Kosovo, though the majority fled.

Though an aerial bombardment has an effect of rallying the people around the leader, losing a war does not have the same effect. Kosovo was the heart of Milosevic's nationalist politics. Posing as the protector of the Serbian minority there is what brought him to power, and failing at this will have had an impact on the elections. The credit for deposing Milosevic goes to the Serb people, but the Kosovo war will have played a role in it.

The Serbian economy recovered to its pre-war level in 2002 or 2003 and has since grown by 7% in 2004 and 5.9% in 2005.

The lawlessness and banditism in Kosovo was already there before the war began. In addition there was a guerilla war, which was threatening to spill over into Macedonia.

In total, I don't think it is arguable that the overall situation now is not better now than before the NATO intervention. What parts of the overall improvement were caused by the intervention can be argued about, as well as whether the intervention was worth it and whether the situation could not still deteriorate.