COVID-19 Impact: Some Contrarian Views

Two important questions: How deadly is the virus and what should be the policy response. Two articles that make convincing arguments about deadliness of the virus being overestimated are as follows:

1. Covid-19 total lockdown: An economic and humanitarian disaster by Rupa Subramanya on orfonline.org

2. Is the Coronavirus as Deadly as They Say? by By Eran Bendavid and Jay Bhattacharya in the Wall Street Journal.

Both the above articles argue that the actual fatality rate of COVID-19 should be much lower than what is being reported if calculated correctly. The second article is behind a pay wall unfortunately but the crux of the argument is based on the wrong way of fatality rate calculation- should be calculated as percentage of infected population and not as of confirmed cases done by several epidemiological modelers. Here is what they say:

How can we reconcile these estimates with the epidemiological models? First, the test used to identify cases doesn’t catch people who were infected and recovered. Second, testing rates were woefully low for a long time and typically reserved for the severely ill. Together, these facts imply that the confirmed cases are likely orders of magnitude less than the true number of infections. Epidemiological modelers haven’t adequately adapted their estimates to account for these factors.

Given this background, an article by two renowned development economists, Debraj Ray and S Subramnain becomes especially pertinent.

In India’s battle against Covid-19, we are inevitably confronted by the choice between social distancing on the one hand, and denying people their livelihood on the other. Recognising the unsustainability of a general, mandatory lockdown, Ray and Subramanian put forth a proposal whereby the young are legally permitted to work and the locus of measures to avoid intergenerational transmission is shifted to the household.

I am glad that at least the Delhi and Kerala governments are resorting to humanitarian policies like opening food shelters to care for the plight of migrant workers, however, it might be worth deliberating on the policy proposals by Ray and Subramanian above.

Dynamic macro perspectives on India’s Monetary Policy

I reviewed the book Monetary Policy in India: A Modern Macroeconomic Perspective edited by Chetan Ghate and Kenneth M Kletzer, Springer, 2016 for EPW recently.

https://www.epw.in/journal/2018/28/book-reviews/monetary-economics.html

Overall, the articles in the book represent a significant and substantial contribution to the literature on monetary policy of India- definitely from the point of view of application of dynamic macroeconomic models to the policy environment of a typical developing economy.

 

NPR’s Podcast on Demonetization

NPR’s Planet Money did a two part podcast on Anil Bokil and his organization Arthakranti who convinced Prime Minister of India Mr. Narendra Modi about demonetization as a solution to India’s poverty, corruption, and black money related problems. This was back in 2013 when Mr. Modi was Chief Minister of Gujrat.

Arthkranti is a plan conceived by some accountants and engineers to curb corruption and improving tax compliance in India. According to them, limiting the use of cash is one of the solutions. I have already talked about what is wrong with this solution here, here, and here. I also wrote a critique of Arthakranti proposals few years ago while I was grad student and did not think that anyone would take these proposals seriously!

Nonetheless, it is interesting to hear what Mr. Bokil has to say. NPR guys did a good job here.

I thank J P Konning for making me aware of this podcast through one of his tweets.

On Black Money, Corruption, and Demonetization

Money facilitates more trades and improves welfare than what is possible without it. Monetary theorists would call this as money being ‘essential’ because the total volume and value of transactions achievable with money is much bigger than the one achievable without money (see pp. 47, Nosal and Rocheteau, 2011). From this perspective, demonetization of November 8, 2016 definitely reduced economic well-being of Indian people at large overnight. However, the effect may not be just this as temporary one-time reduction in the achievable set of transactions but also the ones in immediate future. While current markets in goods and services facilitate current consumption and investment, credit markets allow economic agents to smooth production and consumption over time. A pervasive reduction in liquidity therefore, however short term, is bound to adversely affect both current and future consumption and investment decisions. This effect could be pronounced in case of the Indian economy where a huge proportion of transactions are in cash including cash used for lending through informal channels like money lenders as well as microfinance. Accordingly, it should not be a surprise if there is a significant reduction in real GDP over at least 3-4 quarters if not more.

So what was the objective behind the policy? Initially, we were told that the main objective was to curb corruption, terrorism, and black money. However, the narrative slowly changed as days passed and the outcome discussed now is making India a cashless economy. Accordingly, there are two separate sets of questions that need to be asked. They are as follows:

1. Is demonetization the best way to deal with corruption and black economy?
2. Is there something inherently good about a cashless economy over a cash based economy that makes it a desirable goal?

Let us look at these one at a time.

Dealing with corruption and black money:

In order to figure out how to curb corruption and black money, we first need to know what gives rise to these phenomena. If we do not know that then treating these problems by demonetization would be akin to treating a person with high blood sugar by replacing all the sugar laden blood with fresh healthy one without any thought to what lowered sugar absorption in the first place!

It turns out that almost all the countries have some degree of black economy (also called shadow or informal economy) and hence there has been significant amount of research on its causes and determinants. According to a survey of such literature by Schneider and Enste (2000) , the most important and often cited causes are:

1. the rise of the burden of taxes and social security contributions;
2. increased regulation in the official economy, especially of labor markets;
3. forced reduction of weekly working time;
4. earlier retirement;
5. unemployment; and
6. the decline of civic virtue and loyalty towards public institutions combined with a declining tax morale.

Clearly, all of the above reasons seem to be relevant for India with varying degree of intensity giving rise to the black economy. Even if demonetization reduced the existing stock of black money, a different set of policies would be required to reduce its growth in the future.  Research substantiates these claims. For example, studying the old and new EU member states, Fialova and Scneider (2014)  indicate that one institution that unambiguously increased shadow economy is the strictness of employment protection legislation. Besley 2004 shows that the states in India that implemented pro-worker laws experienced lowered output, employment, investment, and productivity in registered or formal manufacturing and an increase in output in unregistered or informal manufacturing. Pro worker laws were also associated with increase in urban poverty. So in order to reduce black economy one of things that the government could do is reform the labor laws that currently make hiring labor through formal channels costly. Now that is a tough decision with significant political cost makes it a less attractive option for any political party in power.

Other regulations that introduce opaqueness in implementation of and compliance with them also increase the size of black economy. The policy of GST is right step in this regard and so would be streamlining of other direct and indirect tax laws and regulations. India has consistently fared worse on the ‘ease of doing business‘ index and one of the main reason is the lack of transparency in rules and regulations governing business transactions. Reforms in these laws would certainly reduce the growth of informal or black economy.

I think India suffers from a low tax morale because people do not trust the government to deliver on public goods and hence avoid paying taxes. In response to poor quality of public goods, it is natural that people would hoard cash to solve the lack of public goods problem for themselves. For example, some people may evade taxes because of high cost of private education for their kids, the necessity of which arises because of serious shortage of affordable public education. If people cannot meet the needs that should be satisfied by public goods, then they are more likely to engage in corruption in addition to tax evasion making corruption and black economy complementary goods. This is not mere speculation but something that research substantiates. For example, analyzing a sample of 51 countries around the world over the period 2000 to 2005, Buehn and Schneider (2011) present empirical evidence for a complementary (positive) relationship of corruption and the shadow economy.

In short, the set of policies that could influence the size and growth of black economy as well as corruption have to do with transparency of policies governing different economic transactions in the economy. Unfortunately, demonetization is not one of them.

Desirability of Cashless Economy:

Let us go back to the way monetary theorists think about use of money in its various forms in the economy. As mentioned above use of money expands the size of economic pie. What if we still use money but now only in its electronic form- that is we go totally cashless? Is that desirable from a theoretical point of view? We will use the same logic as before: look at the set of transactions that use of cash makes possible and compare that to the one made possible with use of electronic money. Clearly everything that can be done through use of actual cash can be done by use of electronic money. There will be some kinks to smooth out- for example the infrastructure required to go totally cashless should be present and all that- but let us for a moment assume that all the preconditions to go cashless have been met. Would it lead us to a better size of the economic pie or bigger set of allocations compared to what is possible with just cash? From a pure theoretical point of view, the answer is no because actual cash has an important advantage over its electronic form- possibility of anonymity!

For transactions that could be completely legal, but out of privacy concerns I may not want them recorded. Secondly, may be the value of the transaction may not be high enough to justify the cost of using the electronic payments infrastructure. If we move from predominantly cash based economy to the one just based on cash in electronic form, all such transactions would be difficult and that will reduce the size of economic pie or even worse push such transactions to the underground or black economy. Therefore, the policy of going cashless will most likely give rise to a different kind of black economy or alternative hard to trace forms of payments like crypto-currencies.

To make use of electronic payment systems economical, commerce would have to organized in a different way. The smaller neighborhood grocery shops may have to give way to larger ones to benefit from economies of scale. This might be easier said than done because these small neighborhood grocery stores also provide credit to their customers. Because each of these stores deal with a smaller set of consumers, it is easier for them to monitor repayment. This ability of smaller grocery stores to provide and monitor credit allows credit constrained consumers to manage the mismatch between expenditure and income flows. If we want people to shop in bigger stores instead, then we will have to provide people access to formal credit markets. The presence of informal economy makes verifying income flows difficult and therefore access to formal credit markets limited. Hence, reduction in the black economy would make it easier to increase the use of electronic payments rather than the other way round.

Policies to reduce corruption and black economy:

We are back to the question again-what are the policies that would reduce the extent of black economy? A preliminary look at the data tells that tax evasion, corruption, and black economy are problems more acutely present in developing countries than in developed ones. This suggests there is something more fundamental than availability of high denomination currency notes that is causing this. Besley and Perrsson (2014) provide some answers and they are closely related to the importance of institutions argued very convincingly in Acemoglu and Robinson’s Why Nations Fail. Economic institutions are the set of rules that govern our economic transactions and outcomes. Most developing or poor countries have extractive institutions that allow access to resources and economic opportunities to only few and often at the cost of the majority. They do not provide clear property rights making it difficult to claim returns to investment in physical and human capital. We can see clearly from the analysis above, corruption and black economy are a result of extractive institutions. For example, poor property rights also make it easy to hide unaccounted cash in real estate. So to reduce black economy, we need to replace the extractive institutions with ones that broaden access to resources and opportunities and help build capabilities. Such institutions are called inclusive institutions examples of which would be impartial rule of law, clearly defined property rights, access to public education and health care, and well-functioning markets. Unfortunately, there are no shortcuts- and demonetization is not even that- it is actually stripping people of their rights over their income making it an extractive institution.

What can scrapping larger currency notes do?

Today Narendra Modi, the Prime Minister of India made a surprise announcement to scrap Rs. 500 and 1000 notes- by Nov 8 midnight these notes will cease to be a legal tender. Indian citizens will get couple of months to exchange these notes for the denominations that are still legal tender. The government will introduce new Rs. 500 and Rs. 2000 notes later. According to media reports, the aim of this move is to reduce black money in the economy, tackle potential fake currency allegedly being circulated by Pakistan and curb corruption. The question of course is would such a move be helpful in achieving these objectives.

Would such scrapping reduce the total currency in circulation? It most likely will as people find it hard to convert all their currency hoards in time for denominations that are legal tenders. This one time reduction in money in circulation could help control for inflation and possibly redistribute resources away from the commodities and services bought by such cash. However, this effect might only be temporary. The new Rs. 500 and Rs. 2000 notes will become the new store of value for people who have a need to hoard and transport large amount of cash. The time period between abolition of current high denomination notes and introduction of new ones will be the time when hoarding black money would be difficult. But ultimately, if the government thinks that black money is a function of availability of high denomination currency notes then not much will change in the long run. To make sure hoarding and transportation of cash is difficult on a more long term basis, it might be advisable to reduce high denomination notes permanently and possibly have it done simultaneously by other countries as argued in this recent article from the Economist.

To address the problems of black money and corruption on a more permanent basis, we will have to look at root causes of these phenomenon. Not all black money is illegal- the monthly cash transaction between you and your maid is also a part of the black economy. Every time you decide not to insist on a receipt for your transaction, you are participating in the black economy. Hundreds and thousands of Indians participate in these transactions on a daily basis contributing to the black economy. So behavior of common people is a part of the problem and solution. There is also an element of redistribution in corruption. If there is lack of equal opportunity and/or presence of persistent inequality, corruption might serve as a way of redistributing money from the haves to the have nots. Sometimes black economy might be a reaction to excessive state regulation and corruption. There is a rich literature that looks at the causes and consequences of shadow or informal or black economies. See Schneider and Enste (2000),  a comprehensive survey published in the Journal of Economic Literature.

Therefore to seek a long term solution to problems of black money and corruption we might have to dig deeper than just surprise scrapping of higher denomination notes.

Update: Read my critique of the Arthakranti proposals here. Looks like some of the ‘wisdom’ behind Modi’s move came from that!

On Cows and Central Tenets of Capitalism!

Santosh Anagol has been doing interesting research on several phenomena concerning the Indian economy. In a recent paper, he and his coauthors estimate that returns to owning a cow in India are negative and hence the continued existence of cows violates the central tenets of capitalism.

Daron Acemoglu and James Robinson discuss these findings in their extremely interesting blog here. They argue that social embeddedness could help us understand why cows are still a part of a typical Indian farmer’s portfolio. I think that their arguments certainly makes sense, however, one might also look at new monetarist economics for an answer to this question. For example, Lagos and Rocheteau (2008) analyse an economy with money and capital as competing media of exchange and their model I think could explain this puzzle that Anagol et.al pose.

In their economy, agents over-accumulate capital in a non-monetary equilibrium because the capital asset performs the function of a productive asset as well as a liquid medium of exchange when needed. The introduction and use of money therefore allows the liquidity use to be separated from the productive use and corrects the inefficient over-accumulation of capital. Thus, fiat money plays a welfare enhancing role in this economy. However, that precisely does not seem to be happening in India and that is the puzzle that Anagol and his coauthors are referring to.

So why do Indian farmers seem to be preferring to invest in an asset that has negative returns, despite the availability of a liquid medium of exchange? Here, it is important understand what constitutes return on capital. Lagos and Rocheteau propose that return to capital in such an economy can be thought of being comprised of two parts: a liquidity return referring to capital’s role in exchange process and the intrinsic return associated with the productive use of cows. Anagol’s analysis seems to be capturing only the intrinsic return while cows continue to have a liquidity return (premium) in the minds of Indian farmers. Ideally, this perceived positive liquidity return for cows should not prevail if fiat money does provide the necessary insurance against uncertainty. The fact that it does implies that the insurance provided by access to fiat money is not enough.

Anagol and his coauthors do raise this point but dismiss it citing the proliferation of different forms of microfinance institutions in rural India increasing access to savings. However, research has shown that actual use of these institutions is quite uneven and tends to depend on factors that could be explained using the economics of networks. For example see Matt Jackson’s work here. I think the factors mentioned above still continue to influence the basic uncertainty that farmers face in a substantial way. I am not sure if microfinance would be able to provide enough insurance in case of a crop failure for example. Because provision of funds in such case may require access to a mechanism to transfer funds from non-affected areas to affected areas to meet the demand and microfinance in its current state most likely is not in a position to handle that.

I still think that social embeddedness plays an important role. On the one hand traditional socioeconomic relationships have broken down reducing the access to mechanisms that could serve as partial insurance mechanisms and on the other hand access to modern monetary economy is still hard to come by. Lack of roads, absent storage and refrigeration facilities, ineffective or absent land reforms, inadequate irrigation facilities keeping agricultural output sensitive to rainfall shocks, all imply that the benefits from participating in the market economy only add to the existing uncertainty that these farmers face. Till these issues are addressed Indian farmers will continue continue to hold cows despite their negative intrinsic return.

On Powerful Macroeconomic Concepts: Consumption Smoothing

The idea of consumption smoothing is a very powerful one and seem to underlie a variety of economic as well as social phenomenon. In what follows I discuss a few examples to illustrate this. Consumption smoothing implies that the people prefer a smoother consumption path over a relatively choppy or fluctuating one. The ability to smooth consumption differs across countries and within countries across income classes. For example it has been documented that private consumption expenditure in developed countries is less variable than the real GDP at the business cycle frequencies, while in the developing countries it is more variable than the real GDP. This difference can be explained by differing access to credit markets as well support from governments in terms of welfare spending. You can read more about this here and here.

A couple of research papers on India also highlight how consumption smoothing can help explain patterns of migration and marriage as well as the probability of survival of a girl child in rural India. For example, Rosenweig and Stark (1989) find evidence among rural households in India that marriages are arranged between families that come from areas with different income risks. This allows for inter-household transfers of good and services and helps households to tide through rainfall shocks. Such arrangements are important in the absence of formal agricultural insurance products and difficulties in credit provision. Rainfall shocks are not uniformly distributed over the region and hence such flow of goods and services associated with such marital arrangements helps families smooth consumption in difficult time periods.

In another study on India, Elaina Rose shows that one can explain excess female mortality in rural India with the help of consumption smoothing. Using data from almost 4000 rural households, Rose finds that the ratio of probability that a girl survives until school age  to the probability that a boy survives to this age is related to rainfall shocks in childhood. This ratio shows improvement for a cohort that experiences a positive rainfall shock in the first two years of life.   This means that when there is an increase in income or purchasing power of a household because of good rainfall, the probability of a girl child surviving improves as households can afford to allocate more resources to the girl child. The opposite happens when there is a negative rainfall shock. The general  importance of rainfall shocks in fluctuations in Indian GDP is discussed here.

However dismal these findings, they have important implications for government policy. As Rose argues, becuase excess female mortality in rural India is associated with inability to smooth consumption  through other means, promoting institutions that provide alternative mechanisms to smooth consumption might do far good in improving survival of the girl child than any other policy. In case you doubted the importance of insurance and other financial products that provide a hedge against income fluctuations, you will find som solid argument here!

 

To deregulate or not?

An article published at Macroscan on the dilemma of the Indian government about whether to deregulate the price of oil or not, the author argues for not to do so. However, I think he needs more than the analysis he is basing his argument on. The simulation is based on Tinbergen style simultaneous equations model of the Indian economy. So my 5 cents to the debate are as follows:

There are several general equilibrium effects of an oil price shock that have to considered. How are people going to react to the change in price of oil? In the first place, shielding the consumers from oil price shocks has distorted consumer decisions. Combined with shoddy public transportation system, it has lead to a higher demand for private transportation vehicles. If government passes on the oil price changes to the consumer, the consumers might respond to the relative price changes. Over the period of time there will be further demand for efficient public transportation and the reliance on oil for private transportation might actually go down leaving the net effect on GDP close to zero. Also, reduction the oil subsidy will reduce the over fiscal deficit and lower the inflation tax. The government might decide to channel that expenditure somewhere else like better schools or highways! But to account for such kind of changes you will have to simulate a micro-founded general equilibrium model and not a Tinbergen style simultaneous equations one (NIPFP working paper No. 2012-99) which does not allow for equilibrium responses from economic agents (the famous Lucas critique!). I think right assessment of what should be the appropriate policy in the case of oil price deregulation cannot be made till such analysis is undertaken. You still might have a case for not deregulating the price of oil but it would be based on a more robust analysis.

Factor shares in India’s National Income

The business cycle properties of data in the US says that consumption is much less volatile than the GDP. This suggests that households do engage into consumption smoothing and hence looking at consumption distribution is not a good gauge for what is happening to income distribution. A similar argument can also be made for India and hence the debate on effects of liberalization policies would do better if based on income distribution than just on consumption distribution. C.P. Chandrashekhar and Jayati Ghosh make this point quite well in their recent column in The Hindu Business Line.

According to their analysis the share of wages and salaries in the national income of India has shown a decline since 1991. This decline is evident both as a share of total NDP as well as of Organized sector NDP. It was roughly around 70% for a decade preceding the economic reforms and has declined since to 50% in the year 2009. This might seem surprising given that in the US (and probably most of the developed world ) the share of compensation of employees in national income has remained between 60-70% for last 50 years or so.

The authors suggest this as an evidence for rising income inequality after the economic reforms and I don’t necessarily disagree with that interpretation.This issue is certainly important to look into and might suggests a role for policy intervention.

However, the contrast with the US suggests that there might be some other factors at play causing the shares to settle at different values in both these countries. One reason for this contrast is that the factor shares could reflect the relative factor scarcity. Capital being relatively scarce in developing countries compared to the developed ones, higher overall returns for it might be expected. The other reason might be the declining importance and presence of unions in the Indian organized sector after reforms than before. If one admits that most of the growth of the organized sector has been because of the rising service sector, then this does makes sense. In addition, the continuing rigidity of labor laws might also mean a lower opportunity cost for ones time further reducing the bargaining power of the workers.

Overall, these empirical regularities and differences in factor shares across countries are definitely worth investigating more.

Update- January 24, 2014: The recent issue of QJE has a paper on this issue. Looks like declining labor share is not just an Indian phenomenon. The authors surmise that the relative decline in price of investment goods explains this trend. You can read it here.

Economic Oportunities and Fertility Behavior

While Indian policy makers attack the population growth problem using variety of incentives, it looks like the economic incentive trumps all others. In a recently published paper in the Quarterly Journal of Economics, Robert Jensen finds that increased awareness of prospects of gainful employment made the women in the treatment village to be less likely to marry and have kids than those who were not aware of such opportunities (control village). Those married were also more likely to limit the number of kids in order to pursue a steady career.

The economic reasoning behind the behavior displayed by these women is simple. An increase in the economic opportunities increased the opportunity cost of getting married and having kids leading to a substitution away from them. A good example of rational behavior from rural India. You can find the paper here.

Reference:

Robert Jensen, Do Labor Market Opportunities Affect Young Women’s Work and Family Decisions? Experimental Evidence from India The Quarterly Journal of Economics (2012) 127(2): 753-792