Ricardian equivalence: another reason why fiscal policy works

Sometimes the obvious is hard to perceive. Debate about “Ricardian equivalence” may be missing the obvious: forward-looking, ‘rational’ households should expect fiscal policy to work, and their future incomes to be higher. A Ricardian perspective is therefore supportive of counter-cyclical fiscal policy.*

The idea of Ricardian equivalence is mainly associated with Robert Barro, although it was James Buchanan, replying to Barro, who revived the allusion to Ricardo. The concept is often poorly defined – nearly always assuming what it intends to prove. As a theoretical hypothesis, Ricardian equivalence is perhaps most accurately described as suggesting that households consider future tax burdens when responding to current changes in fiscal policy. Consider a very plausible example: if current state pension provision implies an unsustainable burden in the future due to predictable demographic trends, it is likely that well-informed households will act on the assumption that the rules will change. They are likely to adjust desired savings behaviour or assumptions about likely retirement ages.

Ricardian equivalence is typically cited as a reason why the effects of fiscal stimulus are undermined by forward-looking households. Initially Barro argued that there is no theoretical case for the household sector to perceive its holdings of government bonds as net wealth.[1] He subsequently argues that the ‘Ricardian approach’ suggests that households take the government’s budget constraint into account when reacting to fiscal policy. Frequently he then assumes what he wants to prove: that “equivalence” refers to the idea that forward-looking, rational households will perfectly offset the effects of changing government deficits or surpluses with commensurate changes in their own savings, but with the opposite sign. Oddly, he never considers the significant uncertainty as to the sign of the effect of a change in current budget deficits on future deficits – he just assumes a higher budget deficit today means higher taxes tomorrow.

Ever since the early 1970s, Barro has fought a relentless battle to undermine the efficacy of counter-cyclical fiscal policy. Deficit spending in a recession doesn’t “work” because the private sector anticipates higher taxes in the future, so it raises its savings rate.

Barro has flashes of modesty:

“Most macroeconomists now feel obligated to state the Ricardian position, even if they go on to argue that it is either theoretically or empirically in error. I predict this trend will continue and that the Ricardian approach will become the benchmark for assessing fiscal policy.”

Ironically, I think Barro may be right, but due to empirical and theoretical considerations which are opposite to those he assumes. It seems much more likely that the Ricardian perspective will be used as evidence in favour of fiscal stimulus.

Ricardian equivalence is often characterised, and dismissed, in its hyper-rational form. “Weak” equivalence makes much more sense. It seems reasonable that some households at least may change their medium-term beliefs in response to short-term shifts in fiscal policy. That should not be controversial. What’s odd is the assumption that this negates well-timed counter-cyclical fiscal stimulus. We now have a vast amount of evidence on the effects of fiscal policy on growth. One of the most recent and comprehensive studies of fiscal policy from the IMF concludes the following:

“Our study looks at the experience with fiscal stabilization during the past three decades in a broad sample of 85 advanced, emerging market, and developing economies. The message is loud and clear: governments can use fiscal policy to smooth fluctuations in economic activity, and this can lead to higher medium-term growth.”

If this is the case, why on earth would the private sector raise its savings rate in response to counter-cyclical budget deficits? Fiscal stabilisation raises trend growth. If anything, future taxes are likely to be lower. But more importantly, future incomes are likely to be higher – higher future incomes usually imply lower savings rates.[2] A Ricardian perspective (‘equivalent’ or not) enhances the effectiveness of fiscal policy.

Theory can easily reach the same conclusion. Consider a reasonably sensible, forward-looking firm or household operating in the private sector. In a recession, it’s sensible to assume that a significant proportion of households and firms are credit-constrained (they cannot borrow even if they want to). It’s also reasonable to assume people are myopic (all economics assumes this, by virtue of a time preference discount factor), and it’s probably a rational response to uncertainty. I would go further and argue that many people have an intuitive sense that animal spirits, path-dependence and multiple equilibria matter – without being aware of the jargon, of course. In short, demand creates supply – as outlined in one of the most important papers written since the crisis. So without any empirical evidence, a sensible, forward-looking, household or firm would expect future output to be higher as a result of timely fiscal stimulus.

This line of reasoning is really just the opposite of self-defeating austerity. As Martin Sandbu points out in this excellent piece, a tightening of fiscal policy in Greece would likely reduce debt sustainability. Higher taxes now mean lower future incomes, and quite likely higher future taxes too. A forward-looking, reasonable, Ricardian perspective provides an argument against austerity.

Of course, these observations are very broad brush. I have said nothing about the nature of the deficits, which is extremely important. I have also emphasised counter-cyclical policies, Ricardian equivalence potentially has broader relevance. Simon Wren-Lewis and Jonathan Portes have recently outlined much subtler views on how to run optimal fiscal policy in the medium term. My interpretation of weak Ricardian equivalence highlights the importance of thinking deeply about the effects of counter-cyclical stimulus on trend growth. A combination of measures which work quickly to raise demand (such as cash transfers) with policies that raise the capital stock (infrastructure spending) or accelerate innovation (through lower corporate taxes), address both these issues.

Advocates of aggressive counter-cyclical fiscal policy have missed a trick. Ricardian equivalence is not a counter-argument; its part of the armoury.

 * Thanks to Martin Wolf for triggering these thoughts (to follow him on Twitter: @martinwolf_)

[1] Which is a problem for theory. I don’t know anyone who doesn’t: why else would they hold government bonds? As an aside, it is also clear that as government bonds play an increasing role as collateral in the financial system – i.e. they become more ‘money-like’ – they start to have a significant value beyond their role as a government debt. It is not just an error to view money as a debt, it is also an error to view government ‘debt’ solely as debt!

[2] Studies of the effects of government dissavings on private sector savings rates in the 1980s did indeed show these results – for example the work of Larry Summers and Chris Carroll, although I suspect other forces were at work.