Packing the Courts - excerpts from 'Why Nations Fail,’ by Daron Acemoglu and James A. Robinson

I am in the midst of reading Why Nations Fail, an excellent book on development which attempts to explain why some countries succeed while others do not. Their thesis is that institutions, what economists call the 'rules of the game' the ways in which society is organised make the difference between success and failure. It is a compelling argument but one difficult to summarise but this excerpt from the review in the Washington Post captures some of its essence: 

They are impatient with traditional social-science arguments for the persistence of poverty, which variously chalk it up to bad geographic luck, hobbling cultural patterns, or ignorant leaders and technocrats. Instead, “Why Nations Fail” focuses on the historical currents and critical junctures that mold modern polities: the processes of institutional drift that produce political and economic institutions that can be either inclusive — focused on power-sharing, productivity, education, technological advances and the well-being of the nation as a whole; or extractive — bent on grabbing wealth and resources away from one part of society to benefit another.....

When Congo finally won its independence in 1960, it was a feeble, decentralized state burdened with a predatory political class and exploitative economic institutions — too weak to deliver basic services but just strong enough to keep Mobutu and his cronies on top; too poor to provide for its citizenry but just wealthy enough to give elites something to fight over.

Acemoglu and Robinson argue that when you combine rotten regimes, exploitative elites and self-serving institutions with frail, decentralized states, you have something close to a prescription for poverty, conflict and even outright failure. “Nations fail,” the authors write, “when they have extractive economic institutions, supported by extractive political institutions that impede and even block economic growth.”

But even as vicious cycles such as Congo’s can churn out poverty, virtuous cycles can help bend the long arc of history toward growth and prosperity. Contrast the conflict and misery in Congo with Botswana — which, when it won its independence in 1966, had just 22 university graduates, seven miles of paved roads and glowering white-supremacist regimes on most of its borders. But Botswana today has “the highest per capita income in sub-Saharan Africa” — around the level of such success stories as Hungary and Costa Rica.

How did Botswana pull it off? “By quickly developing inclusive economic and political institutions after independence,” the authors write. Botswana holds regular elections, has never had a civil war and enforces property rights. It benefited, the authors argue, from modest centralization of the state and a tradition of limiting the power of tribal chiefs that had survived colonial rule. When diamonds were discovered, a far-sighted law ensured that the newfound riches were shared for the national good, not elite gain. At the critical juncture of independence, wise Botswanan leaders such as its first president, Seretse Khama, and his Botswana Democratic Party chose democracy over dictatorship and the public interest over private greed.

A critical ingredient are independent courts, that check the power of rulers and help uphold institutions such the rule of law and property rights. Rulers, even well-meaning ones, sometimes find independent courts an obstacle. The books cites different examples of attempts to pack courts that had different outcomes: in the US the attempt was thwarted resulting in a virtuous cycle while in Argentina it succeeded and spawned a vicious cycle. This section reminded me of some recent events so I have reproduced it in full. It's a bit long but I think it's worth a read.  

Packing the court

Franklin D. Roosevelt, the Democratic Party candidate and cousin of Teddy Roosevelt, was elected president in 1932 in the midst of the Great Depression. He came to power with a popular mandate to implement an ambitious set of policies for combating the Great Depression. At the time of his inauguration in early 1933, one-quarter of the labor force was unemployed, with many thrown into poverty. Industrial production had fallen by over half since the Depression hit in 1929, and investment had collapsed. The policies Roosevelt proposed to counteract this situation were collectively known as the New Deal. Roosevelt had won a solid victory, with 57 percent of the popular vote, and the Democratic Party had majorities in both the Congress and Senate, enough to pass New Deal legislation. However, some of the legislation raised constitutional issues and ended up in the Supreme Court, where Roosevelt’s electoral mandate cut much less ice.

One of the key pillars of the New Deal was the National Industrial Recovery Act. Title I focused on industrial recovery. President Roosevelt and his team believed that restraining industrial competition, giving workers greater rights to form trade unions, and regulating working standards were crucial to the recovery effort. Title II established the Public Works Administration, whose infrastructure projects include such landmarks as the Thirtieth Street railroad station in Philadelphia, the Triborough Bridge, the Grand Coulee Dam, and the Overseas Highway connecting Key West, Florida, with the mainland. President Roosevelt signed the bill into law on June 16, 1933, and the National Industrial Recovery Act was put into operation. However, it immediately faced challenges in the courts. On May 27, 1935, the Supreme Court unanimously ruled that Title I of the act was unconstitutional. Their verdict noted solemnly, “Extraordinary conditions may call for extraordinary remedies. But … extraordinary conditions do not create or enlarge constitutional power.”

Before the Court’s ruling came in, Roosevelt had moved to the next step of his agenda and had signed the Social Security Act, which introduced the modern welfare state into the United States: pensions at retirement, unemployment benefits, aid to families with dependent children, and some public health care and disability benefits. He also signed the National Labor Relations Act, which further strengthened the rights of workers to organize unions, engage in collective bargaining, and conduct strikes against their employers. These measures also faced challenges in the Supreme Court. As these were making their way through the judiciary, Roosevelt was reelected in 1936 with a strong mandate, receiving 61 percent of the popular vote. 

With his popularity at record highs, Roosevelt had no intention of letting the Supreme Court derail more of his policy agenda. He laid out his plans in one of his regular Fireside Chats, which was broadcast live on the radio on March 9, 1937.

 He started by pointing out that in his first term, much-needed policies had only cleared the Supreme Court by a whisker. He went on

I am reminded of that evening in March, four years ago, when I made my first radio report to you. We were then in the midst of the great banking crisis. Soon after, with the authority of the Congress, we asked the nation to turn over all of its privately held gold, dollar for dollar, to the government of the United States. Today’s recovery proves how right that policy was. But when, almost two years later, it came before the Supreme Court its constitutionality was upheld only by a five-to-four vote. The change of one vote would have thrown all the affairs of this great nation back into hopeless chaos. In effect, four justices ruled that the right under a private contract to exact a pound of flesh was more sacred than the main objectives of the Constitution to establish an enduring nation.

Obviously, this should not be risked again. Roosevelt continued:

Last Thursday I described the American form of government as a three-horse team provided by the Constitution to the American people so that their field might be plowed. The three horses are, of course, the three branches of government—the Congress, the executive, and the courts. Two of the horses, the Congress and the executive, are pulling in unison today; the third is not.

Roosevelt then pointed out that the U.S. Constitution had not actually endowed the Supreme Court with the right to challenge the constitutionality of legislation, but that it had assumed this role in 1803. At the time, Justice Bushrod Washington had stipulated that the Supreme Court should “presume in favor of [a law’s] validity until its violation of the Constitution is proved beyond all reasonable doubt.”

Roosevelt then charged

In the last four years the sound rule of giving statutes the benefit of all reasonable doubt has been cast aside. The Court has been acting not as a judicial body, but as a policymaking body.

Roosevelt claimed that he had an electoral mandate to change this situation and that “after consideration of what reform to propose the only method which was clearly constitutional … was to infuse new blood into all our courts.” He also argued that the Supreme Court judges were overworked, and the load was just too much for the older justices—who happened to be the ones striking down his legislation. He then proposed that all judges should face compulsory retirement at the age of seventy and that he should be allowed to appoint up to six new justices. This plan, which Roosevelt presented as the Judiciary Reorganization Bill, would have sufficed to remove the justices who had been appointed earlier by more conservative administrations and who had most strenuously opposed the New Deal.

Though Roosevelt skillfully tried to win popular support for the measure, opinion polls suggested that only about 40 percent of the population was in favor of the plan. Louis Brandeis was now a Supreme Court justice. Though Brandeis sympathized with much of Roosevelt’s legislation, he spoke against the president’s attempts to erode the power of the Supreme Court and his allegations that the justices were overworked. Roosevelt’s Democratic Party had large majorities in both houses of Congress. But the House of Representatives more or less refused to deal with Roosevelt’s bill. Roosevelt then tried the Senate. The bill was sent to the Senate Judiciary Committee, which then held highly contentious meetings, soliciting various opinions on the bill. They ultimately sent it back to the Senate floor with a negative report, arguing that the bill was a “needless, futile and utterly dangerous abandonment of constitutional principle … without precedent or justification.” The Senate voted 70 to 20 to send it back to committee to be rewritten. All the “court packing” elements were stripped away. Roosevelt would be unable to remove the constraints placed on his power by the Supreme Court. Even though Roosevelt’s powers remained constrained, there were compromises, and the Social Security and the National Labor Relations Acts were both ruled constitutional by the Court.

More important than the fate of these two acts was the general lesson from this episode. Inclusive political institutions not only check major deviations from inclusive economic institutions, but they also resist attempts to undermine their own continuation. [emphasis added] It was in the immediate interests of the Democratic Congress and Senate to pack the court and ensure that all New Deal legislation survived. But in the same way that British political elites in the early eighteenth century understood that suspending the rule of law would endanger the gains they had wrested from the monarchy, congressmen and senators understood that if the president could undermine the independence of the judiciary, then this would undermine the balance of power in the system that protected them from the president and ensured the continuity of pluralistic political institutions.

Perhaps Roosevelt would have decided next that obtaining legislative majorities took too much compromise and time and that he would instead rule by decree, totally undermining pluralism and the U.S. political system. Congress certainly would not have approved this, but then Roosevelt could have appealed to the nation, asserting that Congress was impeding the necessary measures to fight the Depression. He could have used the police to close Congress. Sound farfetched? This is exactly what happened in Peru and Venezuela in the 1990s. Presidents Fujimori and Chávez appealed to their popular mandate to close uncooperative congresses and subsequently rewrote their constitutions to massively strengthen the powers of the president. The fear of this slippery slope by those sharing power under pluralistic political institutions is exactly what stopped Walpole from fixing British courts in the 1720s, and it is what stopped the U.S. Congress from backing Roosevelt’s court-packing plan. Roosevelt had encountered the power of virtuous circles.

But this logic does not always play out, particularly in societies that may have some inclusive features but that are broadly extractive. We have already seen these dynamics in Rome and Venice. Another illustration comes from comparing Roosevelt’s failed attempt to pack the Court with similar efforts in Argentina, where crucially the same struggles took place in the context of predominantly extractive economic and political institutions.

The 1853 constitution of Argentina created a Supreme Court with duties similar to those of the U.S. Supreme Court. An 1887 decision allowed the Argentine court to assume the same role as that of the U.S. Supreme Court in deciding whether specific laws were constitutional. In theory, the Supreme Court could have developed as one of the important elements of inclusive political institutions in Argentina, but the rest of the political and economic system remained highly extractive, and there was neither empowerment of broad segments of society nor pluralism in Argentina. As in the United States, the constitutional role of the Supreme Court would also be challenged in Argentina. In 1946 Juan Domingo Perón was democratically elected president of Argentina. Perón was a former colonel and had first come to national prominence after a military coup in 1943, which had appointed him minister of labor. In this post, he built a political coalition with trade unions and the labor movement, which would be crucial for his presidential bid.

Shortly after Perón’s victory, his supporters in the Chamber of Deputies proposed the impeachment of four of the five members of the Court. The charges leveled against the Court were several. One involved unconstitutionally accepting the legality of two military regimes in 1930 and 1943—rather ironic, since Perón had played a key role in the latter coup. The other focused on legislation that the court had struck down, just as its U.S. counterpart had done. In particular, just prior to Perón’s election as president, the Court had issued a decision ruling that Perón’s new national labor relations board was unconstitutional. Just as Roosevelt heavily criticized the Supreme Court in his 1936 reelection campaign, Perón did the same in his 1946 campaign. Nine months after initiating the impeachment process, the Chamber of Deputies impeached three of the judges, the fourth having already resigned. The Senate approved the motion. Perón then appointed four new justices. The undermining of the Court clearly had the effect of freeing Perón from political constraints. He could now exercise unchecked power, in much the same way the military regimes in Argentina did before and after his presidency. His newly appointed judges, for example, ruled as constitutional the conviction of Ricardo Balbín, the leader of the main opposition party to Perón, the Radical Party, for disrespecting Perón.

Perón could effectively rule as a dictator.

Since Perón successfully packed the Court, it has become the norm in Argentina for any new president to handpick his own Supreme Court justices. So a political institution that might have exercised some constraints on the power of the executive is gone. Perón’s regime was removed from power by another coup in 1955, and was followed by a long sequence of transitions between military and civilian rule. Both new military and civilian regimes picked their own justices. But picking Supreme Court justices in Argentina was not an activity confined to transitions between military and civilian rule. In 1990 Argentina finally experienced a transition between democratically elected governments— one democratic government followed by another. Yet, by this time democratic governments did not behave much differently from military ones when it came to the Supreme Court. The incoming president was Carlos Saúl Menem of the Perónist Party. The sitting Supreme Court had been appointed after the transition to democracy in 1983 by the Radical Party president Raúl Alfonsín. Since this was a democratic transition, there should have been no reason for Menem to appoint his own court. But in the run-up to the election, Menem had already shown his colors. He continually, though not successfully, tried to encourage (or even intimidate) members of the court to resign. He famously offered Justice Carlos Fayt an ambassadorship. But he was rebuked, and Fayt responded by sending him a copy of his book Law and Ethics, with the note “Beware I wrote this” inscribed. Undeterred, within three months of taking office, Menem sent a law to the Chamber of Deputies proposing to expand the Court from five to nine members. One argument was the same Roosevelt used in 1937: the court was overworked. The law quickly passed the Senate and Chamber, and this allowed Menem to name four new judges. He had his majority.

Menem’s victory against the Supreme Court set in motion the type of slippery-slope dynamics we mentioned earlier. His next step was to rewrite the constitution to remove the term limit so he could run for president again. After being reelected, Menem moved to rewrite the constitution again, but was stopped not by Argentina political institutions but by factions within his own Perónist Party, who fought back against his personal domination.

Since independence, Argentina has suffered from most of the institutional problems that have plagued Latin America. It has been trapped in a vicious, not a virtuous, circle. As a consequence, positive developments, such as first steps toward the creation of an independent Supreme Court, never gained a foothold. With pluralism, no group wants or dares to overthrow the power of another, for fear that its own power will be subsequently challenged. At the same time, the broad distribution of power makes such an overthrow difficult. A Supreme Court can have power if it receives significant support from broad segments of society willing to push back attempts to vitiate the Court’s independence. That has been the case in the United States, but not Argentina. Legislators there were happy to undermine the Court even if they anticipated that this could jeopardize their own position. One reason is that with extractive institutions there is much to gain from overthrowing the Supreme Court, and the potential benefits are worth the risks.

Successful contact tracing and the purpose of lockdowns

Sri Lanka's has experienced a spike in Covid-19 detections since early October and the authorities have responded with a series of localised curfews, shutting down parts of the country, mostly in the Western province. While the centre of the city is now open some suburbs (eg Dematagoda, Borella) have been under curfew for about three weeks while parts of the Western province (eg Negombo) have been under curfew for a month, if not longer.

A lockdown suppresses the disease for a while and slows its spread. While it does not eliminate the disease it buys time while measures to increase public health care capacity (procuring emergency hospital space, breathing ventilators, medical protective equipment, and testing kits) are put in place. It also allows contact tracing to take place.  

"if you can find infected cases, isolate and treat them, and trace the close contacts who they might have infected, and isolate them too, then you can keep much of the infection out of the general population."

This of course is most effective when the caseload is low, if the number of cases is high then the number of contacts to trace multiples which means many more contact tracers have to be employed and much more tests need to be done. Sri Lanka's strategy has been largely about tracing and isolating contacts which has been done with the assistance of the military intelligence and other service units.  

If all contacts are traced and properly isolated then the spread can be slowed or even perhaps stopped completely but this requires specialised skills. 

"Case investigation and contact tracing is a specialized skill. To be done effectively, it requires people with the training, supervision, and access to social and medical support for patients and contacts. Requisite knowledge and skills for case investigators and contact tracers include, but are not limited to:

• An understanding of patient confidentiality, including the ability to conduct interviews without violating confidentiality (e.g., to those who might overhear their conversations)
• Understanding of the medical terms and principles of exposure, infection, infectious period, potentially infectious interactions, symptoms of disease, pre-symptomatic and asymptomatic infection
• Excellent and sensitive interpersonal, cultural sensitivity, and interviewing skills such that they can build and maintain trust with patients and contacts
• Basic skills of crisis counseling, and the ability to confidently refer patients and contacts for further care if needed
• Resourcefulness in locating patients and contacts who may be difficult to reach or reluctant to engage in conversation
• Understanding of when to refer individuals or situations to medical, social, or supervisory resources
• Cultural competency appropriate to the local community"

Public trust is essential for contact tracing to work. People need to volunteer information so that potential cases can be tested and their contacts traced. If people are fearful then they are less likely to visit hospitals or volunteer information. There are a number of reasons why I believe this is low.

First, the worry uppermost in people's mind is that they will be shipped off to harsh quarantine centres with little in the way of facilities. The authorities have supposedly said that people can now isolate at home instead of being sent to a centre but given the lack of consistency in policy how many will believe this?

Second, the sensationalist approach by the media that portrayed this like a cricket match; daily scores of  infections and deaths accompanied by gleeful comparisons with other countries to show that we were 'winning the match'. The public bought into this and seem quite obsessed with the whole issue. 

Third, this approach leads to an 'us' v 'them', mentality, the disease is the enemy and the campaign as a war. This meant that victims became the enemy and were stigmatised. The guide emphasises how patient confidentiality needs to be maintained:

"to protect patient privacy, contacts are only informed that they may have been exposed to a patient with the infection. They are not told the identity of the patient who may have exposed them."

It is the medical professionals who should do the contacting, not the potential patient. In Sri Lanka the authorities paste posters on people's houses warning the public of potential infection which creates stigma while the public helpfully circulates lists of potential contacts leaked from various sources or concocted from somewhere.

Given this environment how many will seek treatment in a hospital for a cough or a sniffle? How many will voluntarily provide information on all their contacts? They may delay treatment until acute symptoms set in.

Undoubtedly the military intelligence are well experienced in extracting information from people but can they get all the information needed? 

The authorities are cremating the bodies of victims but Muslims say this practice is against their religious beliefs. Will this dissuade them from seeking treatment or divulging information? 

For successful contact tracing what percentage of the contacts need to be successfully identified and isolated? I don't have an answer but it is a complex task.

"Another consideration is whether the system identifies most – if not all – contacts an infected person has had during their infectious period. People may not recall (or choose not to reveal) their contacts. It is also not straightforward working out who has had significant contact. The current definition assumes “significant” to mean close proximity (less than 2 metres) and prolonged contact (more than 15 minutes). These are arbitrary thresholds.

Contact tracing is also laborious and time consuming. A study of contact tracers in Sheffield reported that each interview with an infected person to identify their contacts took around 80 minutes. The average number of significant contacts identified for each person infected with COVID-19 – around 30 contacts per case – was also high. Consequently, the workload involved is significant."

The logistics of this become well-nigh impossible if the caseload is large. In sum, even with extensive curfews, the process of contact tracing may fail to identify sufficient numbers of potential victims which means the curfew can only provide a temporary suppression in transmission.

There will be increases in the number of cases following the easing of curfew but the response should be calibrated within the framework of containment, not resorting to repeated curfews. Norman Loayza of the World Bank  warns that:

“Worse than a lockdown is a series of repeated and uncertain lockdowns. The risk of second and third waves of infection is large in populations with low immunity. The prospect of repeated and uncertain lockdowns can devastate the economy and worsen human suffering beyond comprehension.”

    

 

 

The recent increase in COVID cases needs a pragmatic response

Sri Lanka’s most recent outbreaks of Covid are spreading panic. The authorities have reacted with limited curfews and heavy handed isolation; rounding up suspected contacts and bundling them away into inhospitable quarantine camps. Activists have complained of a militarised approach to a public health problem. Similar stories were reported from China. Are the authorities are simply following a Chinese model?

There is also a lot of ire directed towards Brandix, where the first new cases were detected.

Lets try and put the issues in perspective.

People seem to have panicked because they believed the country to be free of the disease. It’s ‘sudden’ reappearance seems to have caused a shock. Were we really free of the disease? I think this was largely a case of wishful thinking. This is not a problem that will simply disappear, however hard we may wish otherwise.

Dr Ranan Eliya of the Institute of Health Policy said as much in March this year and in April warned of community spread.

He also touched the core of the current issue when he said “A competent public health response cannot blame people for its failure – it only indicates the policy itself was badly designed. Blaming the public is an admission of failure.”

What we see now is victims of the disease being blamed for spreading it and being treated as enemies, to be rounded up and imprisoned to keep the public safe.

This approach (even ignoring all the other associated problems) can only work if the disease was eradicated in the first place. Prof. S.B. Agampodi, Professor of Community Medicine, at the Rajarata University also raised the problem of community spread in an interview this week.

“All indicators are that community transmission has been around in the country for some time and it would be very difficult to pinpoint whether the infection started in the Minuwnagoda factory of Brandix Apparel Limited or outside,”

“he explains that when looking at the numbers, there simply cannot be 1,000 positives within the factory in two weeks”.

A response based on short-term knee jerk reactions is not good enough. It’s good politics to ‘show’ quick results but not good policy. This adhocracy is unfortunately only too familiar to observers of economic policy.

The worry is that the government imposes another wide-ranging curfew.

I have wondered before whether lockdowns are an appropriate containment measure for the pandemic in poorer countries. A recent paper (that has not been peer reviewed) also posits that they are not.

One of the aims of a lockdown is to buy time - to allow the capacity of the health care system to be expanded. The problem in poor countries is that the healthcare systems may already be overstretched which means that there is little likelihood of being able to reduce the caseload to a level that is within the capacity of the healthcare system. (ie an already overloaded system cannot cope even with a reduced caseload).

"The healthcare systems in poor countries have few hospital beds and ventilators per capita, and are predicted to be unable to absorb a rapid influx of COVID-19 patients. This means, however, that flattening the curve of the disease to fall within the capacity of the healthcare system may not be feasible, no matter the extent of the lockdown or mitigation efforts employed."

Sri Lanka seems to face this issue, there are only around 500 ICU beds. A recent gift of 200 ventilators from the US has expanded this but capacity seems low. Not much attention seems to have gone into its expansion either.

A lockdown does slow the spread of the disease but the consequence is that the resultant reduction in economic activity results in income losses. Richer countries try to compensate for this with increased welfare payments. Poorer countries with weak social protection schemes are unable to do this, so a greater part of the population suffers.

"Many more workers in poor countries are self-employed or in the informal sector and depend on daily wages to feed their families. In the absence of strong social protection and insurance, the cost imposed by social (and economic) distancing may be large in terms of immediate deprivation and hunger. As a result compliance rates with lock-down orders or social distancing guidelines may be lower in countries with weaker enforcement capacity."

The Labour Force Survey for 2019 (p37) estimates that around 57% of Sri Lanka’s workforce is informal. How did these people survive the curfew? There is no published data but I believe some development agencies did a few surveys and it would be good to have some data to go on.

The few informal conversations that I had with people, I was told that they survived by borrowing money together with what limited handouts they received. The question is, if they are to be subjected to another curfew, will it push them into deeper debt? Could this lead them into a poverty trap? A long term cycle of poverty? The longer the lockdown, the greater the likelihood of debt accumulating. Considering that we already endured eleven weeks of curfew there is a real danger of this. Could another extended curfew condemn a large section of the population to permanent poverty? Policymakers must not ignore this question and in the event a curfew is contemplated adequate relief needs to be provided.

With relief, the government faces two problems – the first that there is no proper mechanism to deliver relief. Samurdhi, highly politicised and poorly targeted, is thought is thought to reach only half the intended beneficiaries. There is a need to build a proper social registry, something the Yahapalanya regime started with World Bank assistance but which is probably yet incomplete. It may not include the whole of the informal sector either. Completing this should be a priority.

The other, perennial problem is the lack of funds – large budget deficits and already elavated debt, both problems made much worse by tax giveaways last year. The solution would be to look to aid, concessionary finance (where available) or resort to money printing, in that order.

In weighing costs and benefits I have been trying to think of this in terms of mortality rates and the costs of livelihood disruption, which is similar to the approach by Barnett-Howell and Mobarak-(but they use a proper model for this, of which more later).

If we look at the mortality rates, the highest in a country not in a war zone seems to be Mexico (10.2%) followed by Italy (9.1%), Ecudor (8.1%). Taking the worst case, assuming the entirety of the population is infected then 10% will die. A prolonged lockdown may save the 10% but it will disrupt the lives of the 100%. Of those disrupted, the 43% in the formal sector will suffer less.

Assuming people in the formal sector keep their jobs (not all will) they will probably be paid some part of their salary (as far as I know only the bank staff and government servants received full pay, others were paid reduced amounts) so they are better placed to survive.

The 57% in the informal sector will suffer more, many may have no income. So to save 10% at least 57% must suffer. If it is a temporary suffering and does not result in permanent increases in poverty perhaps this can be justified. There needs to be a proper assessment of these trade-offs.

It may be an idea to turn to the methodology used by the authors of the paper I mentioned previously. Perhaps either they or someone else could run the models that they used with local data?

To determine the relative value of suppression strategies in rich versus poor countries, we embed estimates of the country-specific costs of mortality developed by Viscusi and Masterman (2017) into the influential epidemiological model developed by the the Imperial College London COVID19 Response Team that predicts mortality from the spread of the virus (Ferguson et al., 2020; Walker et al., 2020)

One reason for the higher mortality rates in rich countries is the higher proportion of elderly (an average of 17.4%) than in poor countries (average 3%). The elderly are more vulnerable so deaths in countries with an ageing population are likely to be much higher than in a country with a younger population, other things being equal.

Thus in poorer countries with younger populations deaths are likely to be less, other factors being equal. Lockdowns therefore will save proportionately less lives in poorer countries.

Sri Lanka stands in-between, it has an ageing society although not as elderly as some of the richer countries.The population over 60 was 12.4% in 2016, which is closer to the older countries than to the younger ones. (For comparison, the elderly account for 22.8% of the population of Italy, 18.3% in the UK and 16% in the US).

The ageing argument therefore not as important a factor in Sri Lanka but for reasons of climate, lifestyle or greater immunity - the BCG vaccine being a possible (but not proven) cause, the death rate does not seem to be very high. Again we don’t have good data, the official statistics being impaired by limited testing but we don't appear to be experiencing a very visible massive increase in deaths.

How should we think about policy for Sri Lanka?

For a start policymakers should have been be gathering data on the costs of the curfew, how livelihoods were affected to assess the costs associated with the exercise. They should also be looking to simple, measures that reduce risks without unduly disrupting livelihoods, the paper suggests some measures:

1. Masks and home-made face coverings are comparatively cheap. A universal mask wearing requirement when workers leave their homes is likely feasible for almost all countries to implement.

2. Targeted social isolation of the elderly and other at-risk groups, while permitting productive individuals with lower risk profiles to continue working. Given the prevalence of multi-generational households, this would likely require us to rely on families to make decisions to protect vulnerable members within each household.

3. Improving access to clean water, hand-washing and sanitation, and other policies to decrease the viral load.

4. Widespread social infl uence and information campaigns to encourage behaviors that slow the spread of disease, but do not undermine economic livelihoods. This could include restrictions on the size of religious and social congregations, or programs to encourage community and religious leaders to endorse safer behaviors and communicate them clearly.