Regressive COVID policies in Argentina
Argentina is moving into its third week of total lockdown. As an economist, I have nothing to say on whether this is the best approach to deal with the COVID-19 threat from a medical point of view, though I know I have a little information on alternatives to be able to evaluate its optimality.
For example, how does the contagion rate change if we are all required to use face masks? Not clear, really. If this number is high (for example, if facemasks greatly reduce contagion) then there may be cheaper alternatives to deal with the pandemic than a total lockdown. But really nobody knows, and it is even politically incorrect to ask.
Analysing sector by sector in Argentina, it is easy to see that only about 35% of economic activity is really shut down … This costs Argentina $500 million per day…
Of course, lockdowns are not always complete. In fact, analysing sector by sector in Argentina, it is easy to see that only about 35% of economic activity is really shut down. But even so, this represents an unprecedentedly large number. This costs Argentina $500 million per day, and if it lasts one month the economy will shrink almost 3% in the year.
Point one: Lockdowns are regressive with respect to workers
Two things come to mind when thinking about the policy responses implemented in different countries.
- My first point is that these lockdowns are probably the most regressive policy interventions ever conceived.
Why? Because lockdowns basically do not allow labour to produce without imposing an equivalent restriction on capital. Thus, workers (imagine informal or self-employed workers) are not allowed to earn a wage, but are obliged to respect all their financial obligations that remain accruing as if nothing were happening.
Why do I have to pay my rent if the government says I cannot use my labour? I believe that is a fair question to ask. A different approach would be to quarantine both labour and capital. Under this approach, during the quarantine there would be no income from labour, but also financial contracts would not accrue obligations. Contracts that were due the first day of the quarantine are moved to the first day after the quarantine, as if no time would have elapsed in between, and so forth.
Lockdowns are probably the most regressive policy interventions ever conceived … [they] do not allow labour to produce without imposing an equivalent restriction on capital … Why do I have to pay my rent if the government says I cannot use my labour?
Needless to say, this requires a change in the legal framework, and such quarantine in contracts will be far from perfect, more so when quarantines are partial. But what the experience of quarantined countries like Argentina is showing is that if you don´t pay heed to this, the contractual renegotiation will come sooner or later, and when it does, property rights will have to be renegotiated and probably will be done so in a more arbitrary and potentially harmful way.
The medical urgency has become an economic crisis and eventually a property rights and an institutional crisis. This is already happening in Argentina: contracts are being arbitrarily redrafted in a chaotic manner. This wave of contractual changes has almost reached the central bank – a draft circulated last week would have weakened its independence further (even if you thought that was not possible) but fortunately was shelved, at least for now.
So my first thought is that a quarantine has to be thought as applying to all factors of production. This will both reduce the anti-labour nature of current lockdowns and, by making the effort more balanced, will reduce some of its contractual and institutional stress.
The medical urgency has become … a property rights and an institutional crisis. In Argentina this is already happening: contracts are being arbitrarily redrafted in a chaotic manner.
The second thought has to do with the almost universal consensus on the need to smooth incomes throughout this crisis (see chapters in Baldwin and Weder di Mauro 2020).
Point two: The consensus to smooth incomes
At first glance, this policy is not controversial; consumption smoothing is at the centre of modern macroeconomics. And the intuition and the policy prescriptions of quickly transferring incomes to the most needy are well conceived. For low-income families, income falls are devastating and need to be compensated (I guess some leaders, such as AMLO in Mexico and Bolsonaro in Brazil, have concluded they cannot get the needed relief to this group, and therefore have thus chosen to coexist with the virus). An additional reason for the need to smooth incomes arises from the need to maintain the employment relationships, which are a relevant capital that is key for a quick recovery as pointed out by Levy (2020).
But now let us think for a minute on, say, the upper half of the income distribution – mostly formal workers and public employees with explicit or implicit long-term labour contracts. The jobs for this group are not at risk, at least unless the pandemic extends for a long time. During the pandemic, many of these workers have no demand for their products, and their own propensity to consume falls.
That this be the result of the quarantine, fear, or solidarity, it really doesn´t matter – the important thing is that they have a smaller desire for consumption.
If we accept that the pandemic changes the utility from consumption, our consumption smoothing intuitions are thrown astray. If utility from consumption comes down (and if this happens, it happens to those that are not on the lower end of the pyramid), the same economic theory that tells you to smooth transitory falls in incomes also tells you that you don´t want to transfer income to states of the nature where the desire for consumption falls … if you do execute that transfer it will have to be paid later on, when those resources will be more valuable.
If we accept that the pandemic changes the utility from consumption, our consumption-smoothing intuitions are thrown astray. If utility from consumption comes down (and, again, if this happens, it happens to those that are not on the lower end of the pyramid), the same economic theory that tells you to smooth transitory falls in incomes tells you that you don´t want to transfer income to those states of the nature when the desire for consumption falls. In fact, if you do execute that transfer it will have to be paid later on, when those resources will be more valuable.
The optimal response
The optimal response will be a combination of:
- a transfer of resources to those that need it and to those whose job relationships are at risk; and
- a sort of unsmoothing for those whose jobs are secure and have seen their consumption propensity reduced.
In fact, it is more efficient to finance, at least in part, the transfer to the first group from the second (more akin to the people to people social protection of Lustig and Birdsall 2020), rather than building burdensome liabilities that will be paid by all in the future.
The optimal response will be a combination of a transfer of resources to those that need it and to those whose job relationships are at risk, and a sort of unsmoothing for those whose jobs are secure and have seen their consumption propensity reduced.
In this moment of dire fiscal need, it seems inefficient to transfer resources to the relatively well off – particularly at a time when they can´t spend. And even if these resources are fully saved by their recipients, it is not clear that the burden of their cost will later on be paid by them.
In this case, an attempt to sustain the income flows for all will not only be inefficient but will increase the regressive nature of the intervention an issue that seems to be missing in current discussions.
Baldwin, R and B Weder di Mauro (2020b), Mitigating the COVID economic crisis: Act fast and do whatever it takes, a VoxEU.org eBook, CEPR Press.
Levy, S (2020), “Suggestions for the Emergency”, UNDP Covid-19 Policy Documents Series. March.
Lustig, N and N Birdsall (2020), “The New Inequalities and People to People Social Protection”, VoxEU.org, 2 April.