>> From a social point of view, I don't see how spending 40% of our wealth on the FIRE sector is efficient.
FIRE is about 20% of U.S. GDP or value added or aggregate income. Now, I do not say that 20% is efficient. What I say is that the growth of FIRE can be broken down into two components: (1) secular trend and (2) departures from trend. Since capitalist production expands over a long period of time, (1) will happen, though at a lower rate than production. Why? Because of the "scale economies" noted by Marx. Cf. the quotations I posted. A certain amount of "liquid" financial assets must exist at any point in time for the M-C-M' value flow to move along the circuit. This reserve of "liquidity" does not contribute to production or profitability. The development of banking (broadly understood) tends to reduce ("minimize," says Marx) the amount of "liquidity" required by the system, thus boosting overall capital's profitability. So, this component is "efficient" from a capitalist viewpoint. Now, to capture this you need to take an average over a prolonged period of time. Because there's also (2) and you need to smooth out such variability.
Now, (2) means that finance will expand -- but that it will also contract! -- about its secular trend. This follows from the boom-bust capitalist pattern of capitalist development. The prima facie case made by the critics of "financialization" is that (2) happened from the 1970s on. Presumably, the current crisis would lead to a recoiling or swing in the opposite direction. The critics would allude to data such as: FIRE in the U.S. used to be 10-13% of GDP just after WWII, yet it is now 20% or so. Since finance is nonproductive, rent-seeking in steroids, etc. (not all the critics of "financialization" are Marxists; "financialization" is not a "vulgar Marxist" or "Marxist" term), then this shows a hypertrophy of these capitalist functions at the expense of capitalist production and capital's profitability. Well, maybe, but not necessarily. I mean, my feeling is that the prima facie case is very strong (I'm playing devil's advocate, so don't shoot the messenger), but I want to see conclusive evidence.
Here's one thing that has to be considered. Nowadays, in a given year, the U.S. "exports" (as does London) a substantial amount of "financial services" to the rest of the world. In that sense, you cannot say, oh, in 1948, US FIRE/GDP = 10% and now 20%, therefore "hypertrophy." No. Because, in 1948, Wall Street was not as much Brazil's, Argentina's, Saudi Arabia's, etc. banker as it is now. The denominator has to be adjusted accordingly. It would not be hard to show that the actual, historical expansion of capitalist production in, say, China or India in the last 40 years was made possible (in part) by the growth of Wall Street or London or offshore banking. Etc. For the "financialization" argument to stick, you have to take the broad perspective and show that the growth of *global* banking in the last 40 or 50 years outpaced the growth of *global* capitalist production or that it could have been significantly more modest without affecting negatively the average profit rate and the growth of global output. Or, using Marx's framework, take the global M-C-M' and show that the overall costs of circulation (or "change of forms") involved over the whole period, measured as a percent of the global capitalist product (flow) or global capital (stock), have increased, and thus pulled down the average profitability of global capital (other things equal). That would be conclusive, since it would show that the cyclical hypertrophy absolutely overwhelmed the secular financial "scale economies" noted by Marx. Circulation costs are not the only factor affecting the average profit rate, but that demonstration would boost considerably the case of those who believe that the current global crisis is a global profitability crisis.
Something like that. The point is that you have to seriously consider the counterfactual, even if it is only hypothetical. Otherwise, the case is -- IMO -- poorly made.