Right now, the IMF appears to some to be on the side of the angels. Their self criticism about the impact of austerity is both unusual and commendable in equal measure. They also appear to recognise that pushing Greece further than it can or should go is stupidity of the highest order. They are producing some of the best policy orientated empirical macro research at the moment (e.g. here and pdf). Yet this is the same IMF that in May co-hosted a conference on Latvia that looked very much like a PR job for the benefits of short sharp shock austerity. (For some links, see here and here.) The IMF’s attitude to austerity remains nuanced and country specific (as noted here, for example).
I’m not going to try and see if those apparently different views can be reconciled. Instead I will offer a more institutional explanation for any apparent confusion. The first point to make is that the IMF is a very large organisation: there are around 150 economists in the European department alone, and that department is one of many: some geographical, some functional (e.g. fiscal affairs, research). These are highly trained, very bright and often very experienced macroeconomists: the shear intellectual fire power can seem overwhelming.
The second point is that, for the moment, there seems to be no strong rigid party line imposed from the top. The country teams that conduct and write the annual Article IV reports appear to have a degree of independence, and so they do not have to completely subordinate their analysis to the political sensitivities of the Board or Managing Director (as was evident for the UK, for example). There is far from complete independence of course: there are mechanisms to try and ensure some consistency of approach, but these mechanisms are not strong enough to achieve total homogeneity.
With so many talented individual macroeconomists, with such a diverse knowledge of particular economies, and in the absence of a strong command and control structure, differences of opinion are not surprising. Indeed, we might expect to see the range of opinion outside the organisation (among academics and policymakers) reflected to some extent within. It would also not be human if country teams were not just a little bit influenced by the dominant macroeconomic perspectives within those countries. After all, the influence of the IMF in most cases is marginal at best, so being attuned to those perspectives is essential.
Having said all this, it might still seem puzzling to find within one organisation such diversity over time as we have seen in the last few years. In 2008/9 the IMF was calling for expansionary fiscal policy to support the world economy, but by 2010 it was arguing for fiscal consolidation. Now the pendulum is swinging back again. One explanation, to possibly misquote a British Prime Minister, is “events, dear boy, events”. In 2009 the panic was the prospect of another Great Depression. In 2010 that possibly had receded, but then the panic was Greece.
A deeper explanation is the argument that the default mode within the IMF has in the past involved imposing austerity on governments that have overspent. So the 2010 switch to advocating austerity was a return to familiar ground. From this perspective it was the championing of stimulus during the worst of the recession that was the more interesting development for the organisation. To quote from the excellent account by Farrell and Quiggin
“The Keynesian resurgence was not entirely a product of the crisis. A Keynesian analysis and associated prescriptions had already begun to emerge in expert debate in January 2008, before the crisis proper hit. Dominique Strauss-Kahn, the Managing Director of the IMF, had announced at Davos that “a new fiscal policy is probably today an accurate way to answer the crisis", prompting Larry Summers to note that “This is the first time in 25 years that the IMF managing director has called for an increase in fiscal deficits" [Giles and Tett, 2008]. Both Strauss-Kahn and the IMF's chief economist, Olivier Blanchard, were pragmatic Keynesians, with a theoretical bent that differed markedly from the previous consensus position at an institution notoriously fond of advocating fiscal retrenchment for countries in difficulty.”
This change in position in 2008/9 probably reflects a pragmatic yet informed response to the unique (after the 1930s) position that the global economy found itself in, with interest rates at the Zero Lower Bound. What is interesting is that the IMF was able to make this change, despite its reputation for promoting fiscal discipline, while other institutions such as the ECB or European Commission (and sections of academia) were less adaptable. Perhaps in comparison the IMF was relatively free of political influence, so basic Keynesian logic was able to win the day, but that would just be speculation on my part. Anyhow, it is good to see this view reasserting itself after the disastrous shift to austerity in 2010.