#66 An Alphabet Soup of Economic Recoveries

The many V-shaped recoveries, 'Act of God' excuse: the citizen version, and the US Presidential Elections 2020

Sep 06, 2020

This newsletter is really a weekly public policy thought-letter. While excellent newsletters on specific themes within public policy already exist, this thought-letter is about frameworks, mental models, and key ideas that will hopefully help you think about any public policy problem in imaginative ways. It seeks to answer just one question: how do I think about a particular public policy problem/solution?

If you enjoy listening instead of reading, we have this edition available as an audio narration courtesy the good folks at Ad-Auris.

India Policy Watch #1: Scarring Effects Of The Pandemic On Indian Economy

Insights on burning policy issues in India

— Raghu Sanjaylal Jaitley and Pranay Kotasthane

What kind of recovery will we have? We have an alphabet soup of opinions. The latest entrant is the K-shaped recovery. It isn’t as strange as it sounds. The argument here is most industries will have a few companies that will benefit from the pandemic in gaining market share and consolidating industry profits with them. So, their graph will rise. While the rest will struggle with debt, lower profits, and an inability to invest. This divergence (‘K’) will be a feature of most industries.

Who knows, really? We are nowhere close to the COVID-19 peak in India. We have no choice but to open up despite the test positivity rates that are high in many states. The consumers aren’t flocking to the malls and shops to spend. Most companies are in a wait and watch mode on capital investments. Any predictions on the shape of our recovery at this moment is a mere shot in the dark.

The officials of the finance ministry and the chief economic advisor (CEA) are confident we are beginning to see a V-shaped recovery. The high-frequency economic indicators are showing an upward movement and there is a belief this trend will sustain.

The Shape Is Irrelevant

We think these discussions don’t offer any insights. Why?

Well, any recovery will be V-shaped once you know the bottom has been reached. Since we know there is no possibility of another full lockdown, we will never see those lows on the graph again. This is simple. So, any curve that has fallen sharply and then begins to rise will look like a ‘V’. The other thing to consider here is the basic unit of time that is used to compare data. The smaller the unit of time, the more the curve resembles a ‘V’. So, if we observe monthly data since June when the unlock started in a meaningful way, we will see a ‘V’ shape emerging. But if we consider quarterly or half-yearly data, the ‘V’ will take time to emerge. So, we expect monthly charts to be in vogue for some time. Lastly, there is also a variable of the length of time you monitor this data. If you show it for six straight months, you can establish a ‘V’ shaped recovery is on and then stop.

We believe there are three questions to ask about the recovery in India:

How sharp is the recovery? Is the ‘V’ narrow or wide?
How long will it take for the economy to recover to its pre-Covid levels?
Will there be some kind of hysteresis or ‘permanent loss’ of economic activity because of the pandemic which will have long-term economic impact? How do we quantify it and, possibly, mitigate it?

Understanding Recovery

Let’s examine these.

In our view, the ‘V’ will be wide. We can’t have a sharp recovery unless the national daily case count falls below a thousand. That looks some time away. And even if we reach there it will take more time after that for confidence to return for economic activities to resume like before. So, if we take a unit of time that’s longer than a month and observe economic parameters for longer than a year, we will find a wide ‘V’ indicating a dull recovery.

The pre-Covid levels will be attained at different times by various sectors. It is clear that travel, hospitality and entertainment will take much longer to go back to their previous levels. Other sectors like real estate, banking and automobile will lag till there’s buoyancy in the job market with real wage growth. Sectors like food, e-commerce, agriculture, IT services will recover the fastest based on evidence during the pandemic so far.

Two points need to be emphasised here. One, there has been a huge impact on the informal economy in every sector. The shift of consumption to the formal economy has accelerated during the pandemic. E-commerce growth is a good example of it. Any growth data emerging from the formal economy must be viewed in this context. Two, the availability of the vaccine and its delivery will be a critical factor in this. Assuming we have most of India vaccinated by the end of 2021, we should see normalcy in 2022.

Considering the above, we shouldn’t fall into the trap of looking at select sectoral data or formal economy trends to figure if we have recovered to pre-Covid levels. We will need to view this at an aggregate level. Even that might be only part of the picture considering how large a part informal economy plays in India. We have suggested earlier we expect 8-10 per cent contraction in GDP in FY 21 (year ending March 2021). A 4-5 per cent growth in the next two years thereafter will mean we return to FY 2020 (ending March 2020) levels only by March 2023. This is a reasonable estimate and our view is it might not be any different for most economies in the world. Our high case count right now might delay us by about six months.

Pictures speak a thousand words. So here are a few projections at how the wide ‘V’s might look like. The key point here is that growth rates can no longer be the only number to look at now. That’s because given that there is a discontinuity in the absolute GDP itself, whenever the recovery starts, it will be from a much lower base. So even a 6 per cent growth rate in FY22-23 will get us to a much lower point in absolute terms than a 6 per cent growth rate in FY19-20 could have. This means the key metric to look at would be the year in which the real GDP could reach the same level as it was in FY19-20.

Based on the former Chief Statistician Pronab Sen’s Ideas for India paper, the chart below indicates when we might return to FY19-20 GDP level (dotted line). The realistic scenario predicts we’ll get there by the beginning of FY22-23. However, that scenario assumed a big fiscal demand push by the government. Given that it is nowhere in the picture still, the pessimistic scenario — where we’ll reach FY19-20 GDP level only in mid-2026 — seems to be getting real by the day.

(Data Source: Pronab Sen, Ideas for India)

Will We Be Scarred?

That brings us to the last point. Will there be a hysteresis or a ‘permanent loss’ in the economy because of our behaviour changing after living through the pandemic. Based on the history of human behaviour after a tail risk event, we think this is inevitable. What we have struggled for the last few months is to find a way to quantify the economic impact of this permanent scarring of our lives.

We had read of various papers that were presented in the annual U.S. Fed (Jackson Hole) symposium that took place online on Aug 27 and 28. One paper that caught our attention was by Kozlowski, Veldkamp and Venkateswaran titled – Scarring Body and Mind: The Long-Term Belief-Scarring Effects of COVID-19. The online version of the paper was released on Wednesday this week. You can read it here. The abstract reads like this:

“The largest economic cost of the COVID-19 pandemic could arise from changes in behavior long after the immediate health crisis is resolved. A potential source of such a long-lived change is scarring of beliefs, a persistent change in the perceived probability of an extreme, negative shock in the future. We show how to quantify the extent of such belief changes and determine their impact on future economic outcomes. We find that the long-run costs for the U.S. economy from this channel is many times higher than the estimates of the short-run losses in output. This suggests that, even if a vaccine cures everyone in a year, the COVID-19 crisis will leave its mark on the US economy for many years to come.”

TL;DR version

This is a rigorous mathematical work that runs into 40 pages with data from the U.S. economy to assess the long-run costs to it. Our summary is below:

Once a tail risk event has been experienced by people or businesses (which are run by people anyway), it stays in their minds. It ‘scars’ them. Tail events trigger belief revisions. Once such an event is experienced, it remains in the ‘dataset of the agent’. The persistence of this memory restrains economic activity long after the event.
The economic impact of the pandemic is modelled as a combination of productivity fall and acceleration of capital obsolescence. The authors have used the tail risk event and epidemiology framework to map changes in our behaviour to their economic costs. The simplified outcome of this is – “whatever you think will happen over the next year, the ultimate costs of this pandemic are much larger than your short-run calculations suggest”.
For the US economy, the model offers the following long-term effects of ‘scarring’ as part of its scenarios (remember these are scenarios, not exact predictions)

A persistent four per cent drop in GDP compared to pre-COVID steady state. Simply put, this is the economic activity lost because of change in behaviour. This is a permanent loss. On a discounted value basis, this is almost 10 times the drop in GDP that may happen in 2020. The reasons aren’t difficult to understand. Businesses will be wary of investing because their risk perception has shifted forever. This reduces the stock of productive capital which in turn impacts labour demand.
On asset prices, the model suggests after the first wave of shock is absorbed, credit spreads and equity will return to their original level. The reasons for this is the firms will cut down on their debt quickly and the discounted interest rates have fallen. The impact of the change in behaviour will be seen in option prices though.
On long-run natural interest rate, also called ‘r-star’, the authors indicate a fall by 67 basis points because of the ‘scarring’ effect

What Does This Mean For India?

The merit of the paper is in its quantification of the long-term impact on economic parameters. The specific quantitative impact may be different (and higher) for India. Our view is because of the duration of our lockdown, the continuing increase in case counts and the state of our economy when it went into the lockdown, we might bear greater long-term costs. For Indian policymakers we have three points to consider from the paper:

We must ensure we avoid capital depreciation or obsolescence. This means credit support to firms to avoid bankruptcy. The long-term costs of bankruptcies because of a shift in belief (of risk) is very high. The usual arguments of ‘creative destruction’ that leads to the efficient allocation of capital and avoids zombie firms can wait for a while. The RBI and the finance ministry have worked in tandem so far to avoid this. We must continue on this path and make sure there is credit transmission happening downstream.
The best time to invest in public health is now. As the authors write:
“If we only updated our beliefs about the ability of a particular type of communicable diseases to disrupt economic activity, then health investments will be highly effective. However, traumatic events often leave survivors with a more general sense that unexpected, disastrous events can arise without warning. This more amorphous fear will be much harder for policy to combat.”
We also need a complete revamp of our public health system. Putting in more money in the current system is like putting lipstick on a pig. Some ideas for reimagining India’s public health system are here in this Takshashila Discussion Document. Of course, there’s a podcast episode on this too.
Lastly, in India, we had the mass reverse migration of workers from cities to villages. This must be reversed proactively. Cities are hubs of enterprise and create enormous value through network effects. A long-term change in perception about cities being unsafe or uncaring for the migrant will extract high economic costs. This isn’t considered in this paper, but it follows from their central thesis of the ‘scarring’ impact. We must work on creating policies that reduce this ‘scar’ about cities than trying to make the villages attractive by providing employment there.

We believe the discussion on economic recovery has to move beyond nomenclature of the shape of recovery to more tangible and real issues on what it would take to accelerate recovery and how-to mitigate the long-term change in beliefs. The narrow sectoral view of recovery, the monthly tracking of charts to see a rebound or the search for green shoots are short-term distractions.

We shouldn’t be missing the woods for trees.

PolicyWTF: No Godly Act for the Citizen

This section looks at egregious public policies. Policies that make you go: WTF, Did that really happen?

— Pranay Kotasthane

Edition#64 briefly discussed how the ongoing tiff between the Union and States on GST compensation might affect local governments. In this edition, let’s look at how this might affect you, the citizen.

If you are late to this issue and want to understand why the Union government’s ‘act of god’ excuse is problematic, look no further than this BloombergQuint piece by Haseeb Drabu, former Finance Minister of J&K. Being an economist who was also involved in the GST Council negotiations, his view is insightful. Key takeaways from his article:

The Union government should not have been made the issue of compensation a ‘centre versus states’ issue; it is not.
Compensation was the clear basis of the final acceptance of GST by states. No ifs, no buts. He writes “The compensation scheme was not designed as an ‘insurance’ but as a ‘minimum guarantee’ of 14% annual growth.”
The usage of the force majeure clause is troubling because this is not a commercial contract but a sovereign guarantee with legislative backing that the Union government is refusing to adhere to.
The compensation cess was not meant to be the only source for funding the state revenue shortfall. In fact, it was added on later to supplement the money available with the Union government for compensating the states. So, citing a shortage in the compensation cess collection cannot be a good reason to deny states their share.

Now let’s get back to the question I began this post with. How does this logjam between two levels of governments affect us? In three ways:

One, given that the Union government has gone back on its guarantee, states are likely to be circumspect of any future bargain deals. For example, bringing petroleum products under GST in the future seems unlikely. This could have brought down prices of petrol and diesel but now the states are unlikely to let go of this cash cow in the near future. In fact, gear up for even higher petrol prices even if the global crude prices are at a low.

Two, the dream of making an unwieldy GST manageable is also likely to remain a dream for a few years. Rationalising the five main rates requires trust between the Union and the states. That seems unlikely for some time now. So prepare to continue to pay higher GST rates. The spectre of the inverted duty structure (which we discussed in edition#50) will continue to haunt businesses.

Three, this will have an effect on India’s social infrastructure. Remember, the constitutional division of responsibilities gives the states the lead role in enabling key social services such as health and education. If states face a revenue crunch, capital expenditures in these areas will be hit first.

All in all, this is bad news for the ordinary citizen. The Union government must consider deferring instead of denying the compensation liability to states, as Haseeb Drabu suggests in his article.

World Politics Watch: An Indian View of ‘Elections 2020’

— Raghu Sanjaylal Jaitley

American elections are upon us. So, it is Tocqueville reading time. Democracy in America, the two-volume treatise of the American democracy and society by Alexis De Tocqueville, is a remarkable work of political and social science. That it was written by someone who hadn’t turned thirty yet makes it a work of genius.

There is a timelessness to its analysis of the American political system, its interplay with the society and the potential faultlines that could emerge from it in future. It is no surprise then Tocqueville is dusted off come election time. There’s always a ready market for any kind of prescience. Democracy in America has never failed to be relevant in any U.S. presidential election since it first appeared in 1835.

What’s Different This Time

Sitting here in India, the U.S. elections this year appear a lost cause for centrists. The usual, lazy way of understanding U.S. elections was simple. The candidates pandered a bit to their base with extreme positions on issues during the primaries. Once the nomination was sealed, they moved to centre and broadened their appeal among the fence-sitters or the swing voters. If you managed this balance well, you created some kind of a new coalition of voters who could serve you well for two terms. The Obama coalition or the Clinton coalition, for instance.

This time it is different. Trump has the lowest approval ratings among any sitting president but he’s wildly popular with his base. Using simple reasoning, he should be softening up on contentious issues to make his tent larger. But he has no interest in doing this. He works to a calculus that’s not easy to comprehend on the surface. On the other side, things are different as well. Biden won the primaries being a centrist. There were so many Democrat hopefuls this time with extreme positions that they cannibalised each other. The centrist space was uncrowded. That worked for Biden.

The Centre Isn’t Safe Anymore

But having won the nomination, Biden is being forced to co-opt extreme positions in his campaign. This is a departure from the norm. There’s a fear the young voters who support extreme causes won’t turn up on election day otherwise. The cheerleaders and opinion-makers of the Sanders-Warren-AOC support base have thrown their lot behind Biden for the moment to defeat the bigger evil that is Trump. But there are caveats in their endorsements and there’s a strong sense they will push team Biden hard to commit to their favoured positions before the elections as bargaining chips for their support. This will force Biden to move away from his centrist position.

Those at the van of the left like Anand Giridharadas have made a strong pitch for Biden with all kinds of strings attached. He writes:

What I hope for Biden is that, recognizing the gravity of this moment, he is able to go beyond the mission of unifying the country, that he will begin to speak of unifying it for the purpose of changing things fundamentally. That he will muster the courage to fight for the institutionalization of his compassion, putting kindness where it counts most: in the law.
And I will say, as an erstwhile skeptic of a Biden presidency, that I believe Biden could, if he chooses, pull it off, because of his great superpower. He has a talent for making the political personal. A generation ago, the feminists taught us that the personal is political. The Biden law seems to be that the political is, and must always be, personal.
…(contd.) A tragic version of a Biden presidency would be an era of good feelings with no reform to show for itself — a situation that would leave America once again ripe for fascism, this time perhaps even with a leader who can read.

The Run-Up Chronicle

If I were a Biden sceptic, those lines aren’t rousing me off my couch to go canvassing for him. Anyway, that brings me to write something I never thought I will write. In some kind of an intellectual way, I think Trump sees through this. Trump understands the psyche of those on the extremes like no other politician in America. It doesn’t matter whether they are on the left or the right. He’s playing out his best-case scenario knowing the odds are loaded against him on how he’s handled the economy and the pandemic response. This is how he expects it will work out.

He will continue to stir the pot on contentious issues and keep his base in thrall. His proxies will amplify the message talking about law and order, the American way of life being under threat and other identity and culture issues. The left-leaning Democrats will hear these messages and turn more strident to counter it. There is never a shortage of incidents involving racial prejudice, police brutality or gun violence in American society. Even sporadic events of these kinds that may happen in the run-up to elections will draw more street protests, violence and looting across states.

How will those of moderate or centrist disposition view these protests? These swing voters are in new territory this time. The candidates aren’t naturally veering towards them like in the past. They have the right-wing cocktail of hate they have contended with for four years on one side and the spectre of leftist anarchy on the other. They will have to make a choice on which is more palatable.

Trump’s Best-Case Scenario

This is what Trump wants the choice to come down to. Biden is forced to move left and defend (or not condemn) what Trump calls ‘anarchy’ or the ‘American carnage’. Trump then has the best of both the worlds. The centrists will compare likely anarchy ahead with four more years of Trump. And the strident left will keep pushing Team Biden away from the centre.

Political or electoral choices aren’t made on what you favour. Instead, they are driven by your deepest fears or anxieties. You vote to avoid them coming to a pass. The great ‘middle’ of America fear anarchy and the upending of their way of life most. That’s what Trump is banking on. He sees the left playing two ‘blinking’ games at the same time. One with Biden and another with those in the centre. Trump is doing everything for the left to win both. Because then Biden will nudge further left and the centre will move to him. That’s his best option now.

Tocqueville’s Fears

So, what about Tocqueville? Have we got anything by him that presages our thesis for these elections? Turns out, we do. While Tocqueville was quite taken in by the social impact of democracy, he foresaw two perils of democracy. He termed them the ‘despotism of public opinion’ and the ‘tyranny of majorities’. He feared these tendencies could lead to two possibilities. One, a subjugation of the wise or the expert opinion to the prejudices of the ignorant who prefer simpler solutions. Two, what Prof. John Keane calls political neurasthenia among the middle class where they are ‘happy to be co-opted or kidnapped by state rulers’ in this essay:

“But what Tocqueville long ago pointed out is that under democratic conditions, especially when the poor grow uppity, the middle class might well display symptoms of what might be called political neurasthenia: lassitude, aching fatigue and general irritability about social and political disorder. Guided by fear and greed and professional and family honour and respectability, they would be happy to be co-opted or kidnapped by state rulers (emphasis ours), willing to be bought off with lavish services and cash payments and invisible benefits that brought them stable comforts.”

Trump is manufacturing an environment that will engender this neurasthenia in the run-up to the elections. A middle class, middle-aged ‘whitelash’ is his best bet. Will he pull it off?

I won’t bet against him.

Share Anticipating The Unintended

HomeWork

Reading and listening recommendations on public policy matters

[Article] T.N. Ninan in The Print on the perils of K shaped recovery in India. He says it won’t be a pretty sight.
[Article] A CNBC article on K shaped recovery in the US and worries over the uneven split between sectors and income groups. An important consideration during an election year.
[Book] Elinor Ostrom’s classic which proved that the tragedy of the commons is not inevitable. Communities can and do self-organise successfully to protect the commons. Nationalisation or privatisation are not the only solutions.

That’s all for this weekend. Read and share.