Some thoughts.. First of all, one of the routine critiques of JKG is that he lacked "rigour", which in economics is identified not so much with quantitative methods as specifically with intermediate analysis or "microfoundations". For non-economists, also known as normal people, these are the juicy bits that explain each step in the process that makes your theory work and are usually left out of explanations intended for non-economists. Although important (after all, the jump from correlation to causation depends on an intermediate mechanism), they often demonstrate diminishing informational returns from quite early in the day. Not coincidentally, as well as being the marker of academic respectability, they also tend to be the least simple to understand.
I don't think JKG was especially interested, having attained the land of tenure, in the admiration of the profession. The reason why his books are worth reading is that they weren't written for professional economists. What he was more interested in was what might be termed a critical theory of capitalism, a way of seeing the economy that gave more place to how the people in it behaved. His distinctive contribution will probably be in the economics of bureaucracy, public and private. Time is said not to have been kind to his`work on alternative models of the firm and The New Industrial State, but I have my doubts.
Certainly, the idea that the decision-making processes of big organisations (in Galbraith's shorthand, the technostructure) were not dissimilar whether those organisations were publicly or privately owned and would have an influence on the economy that grew compared to that of capital might seem to have suffered since the 1960s. That was, at least as a stylised fact, an era of big government, big firms, and big plans. Conventionally, it is assumed that the instability of the 1970s and the conservative turn of the 1980s reversed this trend in a blast of classical competition, as financial deregulation reasserted the power of shareholders.
Well, perhaps. It's hard, though, to see that Microsoft shareholders are exerting much commercial control over the decisions made in Redmond, or that an average Vodafone shareholder is even aware of the terms of roaming agreements, the ETSI standards committees, or the ITSUG patent pool. In a sense, a new wave of firms appeared, grew, and as they grew they took on the Galbraithian characteristics of technocracy. Another post-New Industrial State trend that was justified on anti-Galbraithian grounds was skyrocketing remuneration of top management, notably through the use of share options. This was meant to reconnect ownership and control, but it can just as well be read as the techno-bureaucratic elite managing things for their own benefit.
Another counter-trend has been the effort to make the State's bureaucracies more like private-sector bureaucracies through quasi-marketisation, PFI/PPP, management by targets, and the use of consultants. (There is a small classic to be written about the consultant phenomenon. The divorce of ownership and control has been well-understood almost since Marx, but the recent explosion of consultants represents a further subdelegation of control by the controllers themselves.) At the same time, some types of private organisation have been encouraged to take on state functions, and will no doubt become more statelike as a result.
Interestingly, as they were whipped out of economics like redheaded stepchildren, much of Galbraith's legacy has been welcomed into the security/strategic studies/"defence intellectual" sphere, travelling on a false passport. One of the implications of the technostructure was that it likes a stable environment in order to pursue its goals of institutional growth and technical virtuosity. Hence such phenomena as oligopolistic price stability. Another implication of a multinational technostructure - and the technostructure is nothing if not multinational - is that external political stability is as important as internal economic stability. Galbraith, famously, pointed out that this could be a means of redirecting US/Soviet competition into peaceful activity, and he advised President Kennedy that the space race offered such an opportunity.
Looking at current ideas such as Thomas P.M. Barnett's Core and Gap, John Robb's systems-of-systems, Geir Lundestad's description of the EU and NATO as an empire by invitation, and the "democratic peace", there's quite a strong analogy with the NIS. An integrated, transnational technostructure would not tolerate war or serious political trouble within itself, and would have a common interest in mutual protection. It might even take a benevolent view of those outside it. It would likely be rather dull within, but then, that's what big technical bureaucracies are like.
(Note: My own business, telecommunications, displays the Galbraithian tendencies to excess, which probably influences me. It's not really capitalism, more a planned economy run by an engineer-bureaucratic complex with some competitive areas.)
Update, 08/05/06 1800BST: Our Word is Our Weapon says it better than I can:
Sometimes John Kenneth Galbraith’s writing reminds me more of Joseph Heller than anyone else. He was the closest thing to a satirist economics had (something it badly needed, and still does). I like to think one of his main legacies, like Jane Jacobs who went the same way only a few days ago, will be to have demonstrated the importance of mischief in any form of social science.RIP.