A new research paper by Alesina, Favero and Giavazzi  focuses on measuring the output effect of fiscal consolidations. The idea is that fiscal consolidations tend to have much more favourable effects on the economy if they are done via spending cuts alone, not via increased taxation (see the graph below), which is actually what austerity is supposed to be .
Here's the abstract:
"This paper studies whether fiscal corrections cause large output losses. We find that it matters crucially how the fiscal correction occurs. Adjustments based upon spending cuts are much less costly in terms of output losses than tax-based ones. Spending-based adjustments have been associated with mild and short-lived recessions, in many cases with no recession at all. Tax-based adjustments have been associated with prolonged and deep recessions.” (Alesina, Favero, Giavazzi (2012) "The output effect of fiscal consolidations" Figure 3, pg. 40.)
Tax-based (RED) and Expenditure-based (BLUE) adjustment
The figure shows the results they got when comparing tax-based and expenditure-based fiscal consolidations, using a sample of 17 OECD economies, over 25 years (1980-2005). It is clear that for every country in the sample, tax increases resulted in a negative or stagnant output, whereas expenditure cuts resulted in an increase of output two years after the adjustment.
They have also found (not shown in the graphs) that business confidence picks up immediately after the expenditure-based adjustments, unlike consumer confidence for which it takes longer to recover. Finally, the most important finding, in my opinion, is that the "heterogeneity in the effects of the two types of fiscal adjustments is mainly due to the response of private investment, rather than that to consumption growth". This means that private investments tend to drive recoveries, not consumption as the Keynesians would claim.
The findings of the paper are important in trying to explain the process of "right" or "wrong" fiscal adjustments. Using an increased taxation burden combined with spending cuts is a wrong approach since it depresses the economy temporarily (loss of public sector jobs leads to a loss of consumption before these people are reemployed), but fails to offer incentives for it to grow. All the laid off public sector workers that are supposed to find jobs in the private sector are unable to do so since the private sector isn't hiring due to the many existing constraints it is facing.
Unemployment starts to go up, supported by an increasing number of graduating youths for whom finding a job is now even more difficult, making the situation look bad for the politicians in power. This means that the politicians, still under pressure to close the budget deficit, now need to cease spending cuts and stop firing more civil servants and bureaucrats, since they don't want to make the unemployment picture even worse than it already is. So the government then relaxes the spending cuts and public sector reforms and focuses mostly on increasing taxes to close the budget deficit. The government starts running out of options, as further spending cuts become politically unfavourable while increased taxation is needed to continue closing the budget deficit.
Are public investments the key to recovery?
This got me thinking further on one of the most popular policies aimed at kick-starting a recovery, one that is especially being advocated in the UK  - infrastructure spending. The idea is as follows: the government is suppose to kick-start growth via infrastructure spending by two ways; (1) directly creating jobs, and (2) cutting costs to businesses through improved infrastructure.
The first idea implies that hiring more workers in construction, who will use their new wages to increase spending, will ultimately boost consumption (the classical demand-side story, followed by a Keynesian government spending multiplier). However, if the idea is to hire more workers to start a demand-side increase in consumption, why not have the government hire 100,000 workers (or more), give them all a job to dig holes around the country, and then fill them up? That's a perfectly meaningless job, but as long as the workers get paid, they will begin the cycle of recovery, right? Wrong! Modern consumption theory teaches us  that expectations of temporary income aren't the same as expectations of permanent income.
If people anticipate a rise in income, tax rebates or any other form of stimuli (this is true for companies as well, not just consumers) to be temporary, they will save this money instead of spend or invest it. But when they anticipate a permanent rise in income (like getting a new or better paid job) they are much more likely to spend or invest now as they anticipate a certain future stream of income.
The second way is having the newly build infrastructure lower the costs of businesses and help them grow. But how long does it take to build major infrastructure projects like motorways or railways? A long time! Much longer than the average bankruptcy rate for businesses in the UK. And much longer than the current recovery is going to last (hopefully). Full benefits won't be visible in another 10 years, during which time the very same businesses advocating the project will be using old roads and old railways.
This isn't to say that infrastructure projects aren't important - they are, for long term growth almost definitely, but they cannot be the priority in starting up a recovery. One can argue that major infrastructure projects and things like the New Deal made a big difference during the Great Depression, but the world is much different today than it was 80 years ago. For starters back then information was being distributed either through the radio or word of mouth. In that case the people hearing that the government is ready to do something to help them out was enough to get confidence going. Their knowledge of possible negative effects on public finances was very limited. In the modern age of vast informational availability this is certainly not the case anymore. Particularly among investors, but that's another story.
To start up a recovery, the priority should be on supply-side reforms aimed primarily at resolving labour market inefficiencies and at reducing the regulatory and taxation burden on businesses. These will decrease costs much faster and much more efficient than any new road or rail-link, no matter how good or fast they could be.