by Roberto Perotti
This article appeared originally at Voxeu.org 13 September 2014.
There is a growing consensus that austerity is contributing to the Eurozone’s macroeconomic malaise, but also that spending cuts are needed in the long run to achieve fiscal sustainability. Some commentators have advocated a temporary tax cut financed by unsterilised ECB purchases of long-term public debt, accompanied by a commitment to future spending cuts. This article argues that such commitments are simply not credible – especially given the moral hazard problem created by central bank monetisation of debts.
The consensus is increasing that austerity has not worked – Europe stands on the edge of deflation and suffers from a deficit of demand. A recent VoxEU proposal (Giavazzi and Tabellini 2014) offers a solution that is widely shared on both sides of the Atlantic – all Eurozone countries should cut taxes simultaneously by 5% of GDP, and the ECB should buy the extra debt without sterilisation. This should be accompanied by a credible plan to reduce government spending in the future.
As Giavazzi and Tabellini note, most likely Germany would veto this proposal. But even if Germany went along, I argue that it would not work. To understand why, it is important to recognise that this proposal is one possible form of a ‘helicopter drop’ – a temporary deficit monetised by the central bank, leading to a permanent addition to the quantity of base money.1
The essential feature of a helicopter drop is that it relaxes the intertemporal budget constraint of the state (i.e. the aggregation of the government plus the central bank) – given the path of government spending, and given prices and the interest rate, a permanent increase in base money allows the state to decrease taxes, now or in the future.2 As individuals feel richer after-tax, demand increases and so do prices.
Alternative ways to implement a ‘helicopter drop’
But monetising a tax cut is not the only way to implement a helicopter drop and relax the budget constraint of the state. The same result can be achieved by one version of QE – by issuing base money to purchase existing government debt in the hands of the public. Essentially, the state uses base money (which does not pay interest) to absorb government debt (which pays interest to the private sector). The resulting interest savings can be used to reduce taxes at some point in the future. Yet again, the same result could be obtained by another version of QE – by issuing base money to purchase private assets. The increase in interest receipts by the state can be used to reduce taxes in the future. This could work even if the interest rate on all assets is zero.3
All these actions lead to the same outcome under some, admittedly very restrictive, conditions. To be sure, in practice they are likely to generate different outcomes. For instance, monetising a tax cut today might be more effective in stimulating demand than buying existing government debt and cutting taxes in the future, if there are liquidity constrained consumers who spend all the tax cut today. Monetising a tax cut might also be preferable to buying private sector assets (by crediting the banks’ accounts at the central bank) because it puts money in the hands of individuals and not of banks that might be reluctant to lend it.
However, implementing a helicopter drop with a huge tax cut today is risky, and can backfire badly. The reason is not that fiscal consolidations – the opposite of a tax cut – are expansionary. As I have argued in my recent research (Perotti 2013), there is little evidence that fiscal austerity is per se expansionary.4
The difficulty of committing to future spending cuts
Where is the problem then? Many commentators would agree that in the medium term several Eurozone countries, like Italy or France, need to reduce taxes permanently from their current levels. Virtually all would also agree that a permanent reduction in taxes cannot be obtained with a helicopter drop. The intertemporal budget constraint of the state therefore tells us that government spending must fall permanently.5 A 5% tax cut now might be interpreted as a way to anticipate the benefits of the permanent reduction in taxes, while waiting for the spending cuts to materialise. For this to work, however, one needs precisely a credible commitment to a future path of spending cuts.
Why? Because in the real world, government debt can be and is risky, and markets do not like to see it increase – particularly in countries with a high initial level of debt or spending. Without a commitment to decreasing spending in the future, financial markets might panic, as they would see a return to the loose and irresponsible fiscal policies of the past; and this would hurt the banking sector heavily invested in sovereigns, as in 2011. The attempt to expand aggregate demand via a tax cut would backfire.
The key problem is that in most countries it is impossible to generate a credible commitment to reduce spending in the future, let alone by the staggering amounts required by a tax cut of 5% of GDP. Take the two biggest and most celebrated consolidation plans Europe has seen – Finland and Sweden in the 1990s. Over the period 1992–1996, Finland’s primary deficit should have been reduced by 11.4% of GDP, of which 12.1% in spending cuts; the corresponding figures for Sweden over the 1993–1997 period were 10.6% and 6.8% of GDP, respectively. The IMF took these enormous figures at face value in its recent database on discretionary changes in fiscal consolidations. However, in my own research I have shown that these cuts are based on the announced plans by the incoming governments. The reality turned out to be very different – at the end of the period, Finland cut its primary spending by a mere 0.4% of GDP, and Sweden by 3.6%.
But one need not go so far back into the past. In almost one whole year of work, the spending review initiated by the Italian government in 2013 – without a doubt the most thorough and serious such attempt so far in Italy – has identified at most €10 billion (about 0.6% of GDP) of spending cuts, most of which are still highly controversial and subject to political approval. As of now, nobody knows what fraction will be effectively implemented, or when.
Simply put, credible commitments to large spending cuts in the future cannot be made. The problem is compounded by the fact that the prospect of central bank monetisation creates an enormous moral hazard problem. For those who think this is just a theoretical curiosum, it might be useful to remember that the Italian sovereign debt crisis in late summer of 2011 started in earnest when the Italian government, after announcing that it would cut spending by about €3 billion – just 0.2% of GDP – reneged on its announcement immediately after the ECB started buying Italian government bonds.
One could argue that, if things do not turn out as expected, one can always undo the tax cut. But a country like Italy has never experienced discretionary tax cuts of more than 0.5% of GDP. A swing back and forth of taxes by 5% of GDP would create political mayhem, and enormous economic uncertainty.
Not all deficits are created equal – it is one thing to have a large, temporary deficit in a low-debt country with a history of fiscal responsibility and relatively strong and stable governments – like in the UK after the financial crisis – for the purpose of recapitalising the banking sector. It is a completely different thing to have a large deficit in a high-government-spending, high-debt country with a history of unstable governments and loose public finances, without a credible plan to reduce government spending in the future.
For such a country, a credible commitment to a path of spending cuts far into the future, attractive as it might be in theory, is just not feasible in practice. The only feasible alternative to achieve the overarching goal for the medium term – reduce taxes – is to cut taxes together with spending. This process will take time, and will work incrementally, but it is the only realistic approach. The alternative would defeat its purpose of increasing aggregate demand.
Buiter, W H (2014), “The Simple Analytics of Helicopter Money: Why It Works – Always”, Economics: The Open-Access, Open-Assessment E-Journal, 8(2014-28): 1–51.
Friedman, M (1969), The Optimum Quantity of Money and Other Essays, Aldine Publishing Company.
Giavazzi, F and G Tabellini (2014), “How to jumpstart the Eurozone economy”, VoxEU.org, 21 August.
Perotti, R (2013), “The ‘Austerity Myth’: Gain without Pain?”, in A Alesina and F Giavazzi (eds.), Fiscal Policy after the Financial Crisis, NBER and Chicago University Press: 307–354.
Woodford, M (2012), “Methods of Policy Accommodation at the Interest-Rate Lower Bound”, Paper presented at the 2012 Federal Reserve of Kansas City Symposium at Jackson Hole, Wyoming.
1 Note that Friedman’s original definition of a helicopter drop was different (see Friedman 1969: 4–5) – it consisted of the central bank printing an extra 5% of GDP worth of banknotes, and mailing them to the private sector. The idea was very simple – the extra money base would result in extra money supply via the money multiplier, and this would result in extra nominal demand via the quantity equation. Given velocity, this would either reduce deflation or increase real output, or both. The increase in base money without any increase in the assets of the central bank would reduce its net worth, or the present value of future seigniorage (see Buiter 2014). When the interest rate is zero, there is a problem with the assumption of constant velocity (see Woodford 2012). But more importantly, in the current institutional arrangement, the central bank cannot implement this version of the helicopter drop – to expand the money base it needs to buy assets, hence the other proposal discussed in this article.
2 Note that the two things are not contradictory – a temporary tax cut can generate a permanent addition to the monetary base.
3 If money is indeed irredeemable, buying government debt held by the private sector in exchange for money effectively exchanges a liability of the state for something that is not a liability of the state. This, intuitively, relaxes the state’s budget constraint by more than just the interest saved, and works even if the interest rate is zero. This relaxation of the state’s budget constraint can be used to reduce taxes, hence increasing aggregate demand by the private sector. (See Buiter 2014.)
4 This contradicts the implications of some of my previous research.
5 Obviously, GDP growth and inflation (that eats away some of the nominal debt) would also contribute to satisfying the government budget constraint. But presumably this would not be enough – otherwise a tax cut would be self-financing, and we would have solved all problems in one stroke.