Economics blogs are all atwitter with discussion of Reinhart and Rogoff's (R&R) Excel error: it turns out that a 90% debt-to-GDP ratio is not a critical threshold for dramatically slower growth after all.

All this excitement may be a storm in a teacup but there are wider lessons which go beyond the debate about austerity to the public policy role of economists, with their imperfect analytical tools.

The recovery in advanced countries has been painfully slow. Each is unhappy in its own way, but one common theme has been that, having applied strong fiscal stimulus in 2009 to avoid a second Great Depression, policy-makers then changed tack to focus on the rapidly-growing official debt-burden. Budgets were shifted into austerity mode.

Academic economists provided three possible rationales for austerity: (1) the economy had strong self-righting properties and continued stimulus was unnecessary; (2) austerity would boost confidence and was good for growth (what Krugman dubbed the 'confidence fairy' argument); and (3) debt is close to a critical threshold level and requires immediate action to limit further rises (the R&R argument).

Not much is left of any of these arguments now.

The self-righting properties have been feeble. Confidence is frail, even in those countries which have implemented impressively tough budget restructuring (notably the UK). And now the 90% tipping-point has been demolished.

In any case, none of these one-dimensional austerity rationales was useful guidance for the actual problem at hand: given the slow recovery and the looming debt burden, what was the optimal budget profile over time to steer between these two conflicting constraints? The sensible compromise was to support the weak economy by not unwinding the stimulus quickly while at the same time making firm and credible medium-term plans to get the debt on a downward trajectory. To implement this common-sense approach required detailed judgment, not doctrine-driven imperatives.

It was a different error which was probably more important in encouraging the switch to austerity in 2010. The 2009 stimulus was so successful that advanced-economy growth during 2010 was twice the pace predicted by the IMF. In 2010 it looked like a sustained recovery was underway and serious efforts to rein in debt could and should begin. This was not just the IMF's misjudgment; the OECD and the BIS also saw the recovery as strong and the need for budget restructure as pressing. Meeting in Toronto in 2010, the G20 countries undertook to 'at least halve deficits by 2013 and stabilize or reduce government debt-to-GDP ratios by 2016.' 

But if the forecasts had understated growth in 2010, the next two years were well below forecast, leaving the weak recovery with an austerity setting designed for a stronger environment.

Policy-makers were, at the same time, let down by a more insidious error: an over-reliance on econometric models. At the time, the IMF's model showed that fiscal austerity didn't have a strong contractionary impact. The Fund now accepts that the fiscal multipliers significantly understated the contractionary effect of austerity.

While the Fund has been admirably frank in discussing this revision in fiscal multipliers, it has been slower to incorporate this into policy advice. If the multipliers are bigger, the austerity should be softened if you want to be on the initial forecast path. This altered perception is slowly filtering into the policy commentary. The Fund seems ready to argue with the UK about its degree of austerity. Perhaps the argument needs to be widened to include the US, where growth forecasts are less than half the usual pace coming out of a recession, while the budget is imposing a contractionary impact this year equal to 1-1.5% of GDP.

The IMF is not alone in re-positioning. Larry Summers now thinks the 2009 Obama stimulus was too small, shifting his alliterative advocacy from 'timely, targeted and temporary' to 'speedy, substantial and sustained'. Australian Treasurer Swan agrees, urging less austerity for Europe.

It can be argued that the economists' views, whether right or wrong, were not central to the austerity debate, which took place at a political level. In the US, in particular, it was a doctrinal debate about the proper size and role of government. Economics (eg. the R&R analysis) was just a handmaiden of politics.

But this lets economists off too lightly. Keynes made the point long ago: 'the ideas of economists and political philosophers, both when they are right and when they are wrong, are more powerful than is commonly understood.' Good economists know that nothing is certain and everything depends on everything else. They know the weakness of their analytical tools and resist the temptation to enshrine ambiguities with the false precision of models or with the blinkered certainty of doctrine. Economists have to resist the demands of politicians who want simple arguments ('one-handed economists') powerful enough to confound their opponents and sway the public. Economists need to respond quickly to unfolding events and stand ready to take the defence attributed to Keynes: 'When the facts change, I change my mind. What do you do, sir?' 

R&R's mistake was to draw strong conclusions from muddled data and sloppy technique. They were too ready to accept a result that fitted their preconceptions. If we argue that it didn't matter, then we downgrade the role of academic analysis in policy-making. This leaves the field open to Keynes' 'madmen in authority.'