Jeremy Corbyn's victory has prompted Corbynmania from his fans and talk of the collapse of the party from his critics. Both reactions beg an important question: how much difference do leaders make?

There's a famous quote from Warren Buffett:

When a management with a reputation for brilliance tackles a business with a reputation for bad economics, it is the reputation of the business that remains intact.

What he's getting at is that companies have organizational capital - cultures and ways of doing things - which are very difficult to change. The same might be true of political parties. It is rare for big established ones to collapse, Pasok and the Canadian Liberal party being notable exceptions: the thing about the Strange Death of Liberal England is that it was strange. And it is rare for them to be utterly changed: as Archie Brown points out in The Myth of the Strong Leader, transformative leaders are rare, and require especial circumstances. Those who complain about Blair moving to the right understate his orthodox social democratic achievements in, for example, reducing pensioner poverty and NHS waiting times.

In fact, Buffett is echoing something Marxists have long pointed out, that Labour is fundamentally a social democratic party which has only limited ability to change capitalism: Mr McDonnell's aspiration to transform it might be over-optimistic. One thing Miliband and Poulantzas agreed upon in their famous debate was that there are big constraints upon what parliamentary parties can do. As Miliband wrote:

Social-democrats have tended to be blind to the severity of the struggle which major advances in the transformation of the social order in progressive directions must entail. (Socialism for a Sceptical Age, p163-4)

Let's take just two of these constraints.

One is the danger of capital flight. Any government which tries to raise anything like £120bn from companies - some estimates of the "tax gap" - would see investment collapse; £120bn is equivalent to 30% of UK corporate profits. This greatly limits how much Corbyn can tax firms: in fairness to him, Richard Murphy recognises this.

A second constraint is ideology. Capitalism generates attitudes - cognitive biases - which serve to support the system: the media might exacerbate this process, but is not the sole cause of it. Corbyn will to some extent have to accommodate himself to this fact - and if he doesn't, his successor will. In fact, given his leftist credentials, Corbyn might be better able to do this: only a hardline anti-communist such as Nixon could have opened diplomatic relations with China during the cold war.

In this context, perhaps Yanis Varoufakis has something to offer Corbyn. He, more than most, knows just how difficult the left's job is.

I suspect, therefore that Labour will eventually revert to type, being a moderate successful social democratic party - though the journey will be, ahem, interesting. This offers both hope to the "Blairites" (a word which should be out-of-date by now) and caution to the Corbynites.

That, though, is just a suspicion, tempered by my deep antipathy towards forecasting. What I'm more confident of is that our political debate would be much improved if both sides were to question the ideology of leadershipitis and instead ask: just how much, and through what mechanisms, leaders can change parties? "Cargo cult" thinking is not good enough.