Weekend readings

1. The story of Boeing and the apparent failings in the design of its flight-control software is yet another example of the limitations of outsourcing regulation. Sample this,

The FAA has let technical experts at aircraft makers act as its representatives to perform certain tests and approve some parts for decades. The FAA expanded the scope of that program in 2005 to address concerns about adequately keeping pace with its workload. Known as Organization Designation Authorization, or ODA, it let Boeing and other manufacturers choose the employees who approve design work on the agency’s behalf. Previously, the FAA approved each appointment. Under the new approach, which was fully implemented in 2009, the ODA representatives are still under U.S. legal requirements and the FAA has the authority to oversee them and request that their management be changed. In 2012, a special investigator of the Office of Inspector General at the Department of Transportation sent a memo to the FAA’s audit chief warning him of concerns voiced by agency employees about the new process. Some allegations were made in anonymous faxes sent to the inspector general’s office, and the office followed up by interviewing employees in the FAA’s Transport Airplane Directorate. “Our investigation substantiated employee allegations that TAD and FAA headquarters managers have not always supported TAD employee efforts to hold Boeing accountable and this has created a negative atmosphere within the TAD," according to the June 22, 2012, report sent to the FAA.

The employees told the investigators that managers had overturned a recommendation by staff to remove the administrator Boeing had chosen for the program and “had not adequately addressed employees’ concerns" about potential conflicts of interest, the memo said. The employees, it said, viewed this as evidence of management having “too close a relationship with Boeing officials." Despite those concerns, as well as others raised in a subsequent report by the inspector general, Congress has embraced the program as a way to improve the FAA’s efficiency. President Donald Trump signed into law a change on Oct. 5. It allows manufacturers to request that the FAA eliminate limitations on how company representatives certify “low and medium risk" items, giving them even more authority over their own products.

2. FT on the role of militias in Rio's favelas,

At first, militias offered protection to local businesses at a modest price many were willing to pay. From there to extortion was a short step, and soon militias were selling protection against themselves. They expanded into other services: informal public transport, distribution of cooking gas, pirate cable TV, the sale and rental of commercial and residential property, and more. The most lucrative line of business for the militias has been real estate. Investigators recently found documents at the residents’ association in Rio das Pedras showing that between 5 per cent and 10 per cent of the value of every property deal goes to the local militia. They were also involved in the hugely profitable business of land expropriation or grilagem — the fraudulent assignment of property and land deeds... militias dominated local politics in the areas they controlled... 

During the past 20 years, many new militias have been formed beyond Rio das Pedras. A study last year found they were present in 165 favelas and in 37 other city neighbourhoods in greater Rio, areas of the city that are home to a combined population of more than 2m people. They hand out often gruesome and lethal justice designed to set an example, sometimes for criminal behaviour, sometimes for acts of disobedience such as buying cooking gas from the wrong distributor. Their presence haunts the city

The hardline policies of the new President Jair Bolsonaro is apparently worsening the situation.

3. Talk about regulatory arbitrage - social media platforms edition. How about reaping all the benefits of a publisher without legally being a publisher? This dichotomy is fast becoming apparent as Facebook faces increasing scrutiny over the content posted on its site,

Why on earth do we tolerate technology that can be used to inflame hatred and normalise violence at lightning speed and global scale? The answer lies largely in a 26-word sentence in Section 230 of the Communications Decency Act passed by the US Congress in 1996. “No provider or user of an interactive computer service shall be treated as the publisher or speaker of any information provided by another information content provider,” it states. So the likes of Facebook, Google and Twitter can play by different rules to traditional media companies, including the FT, which are legally responsible for all the content they publish. Jeff Kosseff, author of The Twenty-Six Words That Created the Internet, argues that Section 230 has proved an “awesome benefit” for the tech platforms. It has encouraged astonishing innovation and accelerated the growth of some of the richest companies on the planet. But it has also allowed billions of people to post anything they like online with almost no constraint. Some of that content is inspirational, much of it trivial, and a small sliver grotesque and harmful.

4. Amidst the Varsity Blues scandal of university admission frauds in the US, here comes more evidence of how skewed access to elite universities has become,

at the very best schools, more students come from the top 1 per cent of income distribution than come from the entire bottom 50 per cent.

 5. The latest issue of The American Statistician makes a strong plea in favour of revisiting the interpretation of 'statistical significance',

we should never conclude there is ‘no difference’ or ‘no association’ just because a P value is larger than a threshold such as 0.05 or, equivalently, because a confidence interval includes zero. Neither should we conclude that two studies conflict because one had a statistically significant result and the other did not. These errors waste research efforts and misinform policy decisions.

6. A data journalist at The Economist delves into a few graphs from the magazine and shows how they could be misleading or confusing or failing to make the point, and suggests some alternative illustrations.

7. Finally, US exports of LNG has surged spectacularly since the opening of the Sabine Pass terminal. With another three expected to come online soon, these exports are expected to keep rising and putting a downward pressure on global gas prices.

8. The Economist has this on Chinese locomotives and rolling stock manufacturer CRRC, 

CRRC now employs 180,000 people worldwide and posts annual revenues of $30.6bn, around a tenth of which comes from outside China. Between 2013 and 2017 crrc made 44% of the world’s electric trains and a whopping 71% of its high-speed ones, estimates Maria Leenen of sci Verkehr, a railways consultancy in Hamburg.

The Rise of Finance: Causes, Consequences, and Cures

Finally, my co-authored book with V Anantha Nageswaran is listed for publication on the publisher's website. 

The release formalities in due course and will keep informed here.

Some thoughts on modern development discourse

Consider this parable. Smartman comes to the NewWorld. NewWorld too is populated by human beings who perform basically the same kinds of tasks in their daily lives. But it is far less developed and its people are poor. Smartman tries to make sense of how things work in NewWorld. He is in possession of two instruments - theory and tools for generating evidence (of what works). Smartman completely ignores the reality that the people of NewWorld have deep knowledge about their world.

There is an underlying premise. It goes something like this, "We've been through all this and have reached our present advanced stage of development. So we know what to do. Also you guys have been screwing up so badly for so long. Besides, we bring the tools of scientific empiricism. You better listen to us."

And if the NewWorldman suggests something to improve traffic congestion, pat comes the reply, "How do you know it works? Where is the evidence?". If he suggests that Edtech solutions can do little to improve student learning outcomes in resource constrained and difficult circumstances of public schools, he’s summarily dismissed by the Smartman who argues that the “logical rigour of sophisticated AI-based adaptive learning algorithms can help leapfrog all problems”.

Modernist high-development completely ignores the reality of priors. People in developing country contexts have priors. In any reasonable world, these priors, appropriately validated, should form the starting point for enquiry on important development issues.

Take this defence of the use of RCTs in development. It makes the point that RCTs grounded on theory are easier to generalise since they also contain information about the mechanisms of impact. Point taken. But how many of the big issues in development are amenable to such validation?

Just start with the HIV prevention measure

A randomized controlled trial (RCT) found that showing eighth-grade girls and boys a 10-minute video and statistics on the higher rates of HIV among older men dramatically changed behavior: The number of teen girls who became pregnant with an older man within the following 12 months fell by more than 60 percent.

How does this compare with, for example, enabling access to contraceptives and ensuring its use. Not to speak of development in general and economic growth itself. If someone were to do a regression of various explanatory variables and HIV incidence, one can be certain that just plain development, or increase in median incomes, would be the overwhelmingly largest factor in HIV prevention.

The big questions in ensuring universal immunisation is not about lentils and nudges. It is about availability of vaccines and state's capacity to enable access. Similarly, the primary focus on HIV control is hardly about awareness creation on one particular channel of impact - priming adolescent girls and boys short videos on the higher rates of HIV among older men (sugar daddies). It is about more general awareness creation, strengthening public systems to screen and facilitate rehabilitation and treatment of patients, enabling access to medicines etc. The high-profile RCTs and their nifty results have the result of almost confining the debate on these issues to these RCT-able good-to-have but peripheral concerns. No meaningful dent can be made on the problem with what are at best palliatives. 

There is an interesting psychology behind such articles that seek to examine development. The entire article relies on only two channels of information - theory and evidence from evaluations. What about priors? 

Consider the development landscape in India. The country has had flagship integrated employment guarantee schemes since 1960s, self-employment schemes since 1980s, livelihood programs since 1990s, rural roads programs since 1990s, and nutrition and maternal and child health program for several decades. All these programs have a rich and varied history, with vast troves of implementation data. One would imagine that any meaningful attempt to explore solutions to these problems has to start with a detailed analysis of these historical programs.

Given that the careers of many reputed foreign development economics researchers of today have been made studying the same development issues in India, it should therefore come as a surprise that there is not one paper of note by any of them about the history of India’s numerous flagship welfare programs. Apart from these eminent people coming with a disdain for priors and feeling their theoretical concepts and experimental toolkits are superior, analysis of historical programs would hardly get a place in any decent economics publication or get them tenure in a reputed foreign university.

Not only is history over-looked, it is often sought to be re-written. This paper evaluating the "innovative program" of providing bicycles to girls in Bihar does not find it fit to even mention about the pioneering and very famous bicycle distribution program of Tamil Nadu from the eighties. Or this paper about the use of third party audits in monitoring pollution presents the initiative as an innovation, without any mention of the fact that such audits have become commonplace in engineering works across India for more than a decade. Or this paper's discovery of telephone-based monitoring of the implementation of government programs which conveniently overlooks the reality of the 'innovation' being commonplace across the same contexts. It is as though, in each of these cases, the particular innovation and its assessment started with their papers! Talk about re-writing of history.

Also, in development, the big challenges are more often than not, not about radical paradigm shifts or spectacular innovations, but plain simple good governance and straight-forward implementation of plain vanilla programs and public services.

Doing an RCT to figure out whether third party audits are useful, or telephone call centres are effective in monitoring effective program implementation, or to question the value of hotspot policing betrays arrogance. Who's waiting to be convinced so?

Being outsiders (and thinking that we are more objective, and perhaps even superior beings), we completely ignore the reality that there are real people living in that world and they have sufficient knowledge about their world. About what works, how and why they do. There is a name for observing and studying the world and its people as it exists. But ethnography has long since gone out of fashion.

And as Anand Giridharadas has written, this trend is not just confined to academia, but also in consulting, where observing and listening was once supposed to be the most important traits.

In fact, the irony of the recent interest in system thinking and system transformation cannot be lost. The wheel has turned the full-circle. The only difference being that this fad is just a more impoverished version of rigorous ethnography.

Some truths about impact investing

Development impact investing - social enterprises that seek to combine financial returns with positive impact predominantly on the lives of poor or those at the bottom of the pyramid - has been around for nearly two 15 years. What is its balance sheet?

1. How many predominantly poor focused businesses supported by impact investors have reached $1 bn $100 mn $50 mn in revenues?

2. How many low-cost school or hospital or diagnostic or restaurant or grocery chains have reached thousands hundreds tens of franchises?

3. How many social enterprises have had significant enough impact on hundreds tens of millions of poor people?

4. How many social enterprises have created millions thousands hundreds of productive jobs?

5. How many innovations with meaningful impact on persistent development challenges have been scaled up by impact investors through social enterprises?

6. How many development impact investors have consistently made money outside of micro-finance?

7. How many impact investors have made money in any sector in Africa?

8. How many social enterprises have been able to create a widely-use labour market matching platform for unskilled jobs in any local market? 

9. How many social enterprises have been able to build a commercially viable business out of making available information for different kinds of stakeholders - farmers, traders, fishermen etc?

10. Even on fintech, what innovation has impact investing realised on financial inclusion that has gone beyond the success of microfinance? Give one example of a financial instrument that has scaled  (into becoming a mainstream instrument like micro loans) from fintech that responds to the specific requirements of poor people?

Clearly something wrong with the existing model. Metaphorically, impacting the Bottom of the Pyramid (BoP) requires looking below the submerged ice-berg. Conventional impact investing is all about seeing and searching only the tip of the ice-berg.

Entry costs and business concentration

More from the excellent German Gutierrez and Thomas Philippon, this time along with Calum Jones, on the trends with entry costs and business concentration in the US. I had blogged earlier from Gutierrez and Philippon about how low interest rates has contributed to business concentration and rise in firm surplus 

The reality of business concentration, increase in firms' profit margins, and weaker investment in precisely those industries which have become more concentrated is now widely acknowledged. But its exact causes are a matter of debate. 

Business investment in terms of ratio of net investment to net operating surplus for US non-financial corporates has fallen while that in respect of net buybacks has risen. 

They examine the implied capital gap relative to the Q ratio for the most and least concentrating industries and find that the capital gap is coming from concentrating industries.

The summary of their findings,

Entry has decreased in the US economy, and markets have become more concentrated. We find that entry costs shocks have played an important role and that they are related to entry regulations... our main finding is that time-varying competition has had a significant impact on macro-economic dynamics over the past 30 years. For instance, absent the decrease in competition since 2003, consumption would be 5 to 10 percents higher by 2015 and the capital stock would have been 1 to 3 percent higher by 2015... By 2015, the cumulative under-investment is large at around 10% of capital.

Nigeria facts of the day

From the FT,

Nigeria, a country whose population may this year pass 200m, distributes and consumes about as much electricity as Edinburgh, a city of 500,000... Nigeria’s installed electricity capacity is... around 420kWh a head, or 13GW... Unfortunately, the system can only deliver around 3.7GW to 4GW, or very briefly 5GW on a good day.

And this graphic about government revenues as a share of GDP is stunning.

It is deeply disturbing that government revenues as a share of GDP has mostly declined across the continent over the 2007-17 decade. A reflection of the decline in commodity prices?

Economics for Inclusive Prosperity

There are essentially two types of economists. The first are the garden variety ones, steeped in reductionist models and quantitative analysis, and pretty much ahistorical and single-disciplinary. They create straw-man worlds in which they locate their analysis and suggest baffling or naive solutions. Then there are the old-style political economists, steeped in historical and multi-disciplinary analysis, and engaging with real-world problems and advocating practical solutions.

For me personally, the most frustrating thing about economists in recent times has been the way they have sought to engage with the problem of the alarmingly widening inequality. The mainstream economists of all shades have reduced it to an analysis of the dynamics of different models of economic growth. They have completely glossed over its blindingly obvious real-world impact on the political economy, on the political power balance and capture of the process of making the rules of the game.

This reluctance inability to engage with issues of (political) power imbalances that constrains and enfeebles institutions, lowers the efficiency of transactions and information flows, and raises entry barriers and lowers competition is perhaps the biggest problem with economics today. Constrained by their own limited toolkit and ill-equipped to engage with other disciplines and dimensions of scholarship, they have been a deep disappointment.

In that context, Economics for Inclusive Prosperity comes out as a refreshing change. Suresh Naidu, Dani Rodrik, and Gabriel Zucman outline the excellent new initiative that proposes to use the tools of economics to advocate practical policies that promote "inclusive prosperity" in contrast to the "market fundamentalism" that dominate mainstream economic thinking. The series of essays reconcile rigorous economic thinking with real-world considerations and offer practical solutions for critical challenges that we face today. The underlying principle,

A willingness to subordinate textbook economic efficiency to other values such as democratic rule and egalitarian relationships among citizens. These proposals take Polanyi’s words to heart: to work well, crucial markets (including markets for labor, land, and capital) must be embedded in non-market institutions, and the “rules of the game” must be supplied by government.

A summary,

Arindrajit Dube proposes a system of wage boards, similar to the Australian system, where either administrators or tripartite boards negotiate wages at the industry-occupation-region level, thus setting minimum wages throughout the distribution. He finds that wage inequality would significantly fall as a result. Suresh Naidu discusses the more traditional U.S. labor movement, and how mechanism design, experiments, and behavioral economics can be mobilized to ease the pervasive collective action problem facing unions... Atif Mian discusses the role that inequality, together with capital flows from oil-rich countries and Asia, has played in generating a “glut” of U.S. savings, pushing down the real interest rate and increasing systemic risk. He shows how inequality generates instability in financial markets, but also how private macro-prudential contracting is thwarted because of externalities that contractors are not paying attention to and of specific tax and regulatory structures (e.g., Basel III risk weighting). Exploring the banking sector, Admati shows how banks, uniquely among financial institutions, are overexposed to debt, making them more vulnerable to bankruptcy and a threat to stability. Both authors point to a variety of good regulatory options, with Mian emphasizing credit contract repayments that are contingent on the aggregate state of the economy, and Admati favoring capital requirements and tax reforms that make debt look less attractive...

Gabriel Zucman proposes taxing multinationals by allocating their global profits proportionally to where they make their sales... Sandra Black and Jesse Rothstein use the best modern economics to provide a contemporary restatement of an old idea: government should provide public goods and social insurance... Anton Korinek takes up the increasingly important question of how new technologies affect labor markets and the distribution of income. The direction of technological change is not exogenous, he argues, and it depends on the incentives set both by markets and by governments. In particular, innovators may overestimate the social cost of labor, investing too much in technologies that replace labor... Korinek proposes that they similarly steer technology in the direction of innovations that have desirable distributive properties. They could, for example, promote AI systems that complement and augment the cognitive abilities of workers—along with mechanisms that ensure workers retain a substantial part of the surplus generated. Korinek also discusses how inelastic, complementary factors such as land or specialized skills might be taxed in response to technological change, and how the value of monopolies granted by the patent system is intrinsically inegalitarian since it transfers income from consumers to owners of firms... Dani Rodrik shows that trade agreements ought to include clauses that prevent competition on “unjust” margins, and his “social safeguards” would give countries a claim, justified by broad social support, on trade authorities that a restriction on trade is necessary to maintain the domestic social contract... Ethan Kaplan draws on a few decades of empirical political economy to suggest policies that could drastically alter the balance of political influence in the United States.

This is a great and simple touchstone for any policy making process,

Democratic political economy—where people’s influence on policy is roughly equal and political preferences are arrived at through open, well-informed public debate—must be considered for any policy proposals in 2019. Too many policy ideas break on the rock of government capture by special interests or systematically distorted presentations in the media... 

On power imbalances that are pervasive,  

Many of the essays share the theme of how power asymmetries shape our contemporary economy. Many economists dismiss the role of power because they think it cannot be studied rigorously or belongs outside economics. As Naidu puts it in his essay, “under conditions of perfect competition and information, there is no scope for power.” But asymmetries between different groups abound: who has the upper hand in bargaining for wages and employment; who has market power and who gets to compete; who can move across borders and who is stuck at home; who can evade taxation and who cannot; who gets to set the agenda of trade agreements and who is excluded; who can vote and who is effectively disenfranchised. Some of these asymmetries are traditional political imbalances; others are power imbalances that naturally occur in the market due to informational asymmetries or barriers to entry.

Policies that counter such asymmetries make sense not only from a distributional standpoint but also for improving aggregate economic performance. The policy essays tackle these asymmetries frontally and suggest ways of rebalancing power for economic ends. Unions and wage boards can rein monopsony power in labor markets (Naidu and Dube); putting sand in the wheels of financial globalization can enhance the fiscal capacity of the state (Zucman); regulating private finance can prevent crises (Admati and Mian); giving labor a greater say in trade agreements can improve the design of trade agreements (Rodrik); and restricting campaign contributions and making it easier for poorer people to vote can increase the accountability of the political system (Kaplan).

The impact of QE - Bernanke and Ocasio-Cortez

David McWilliams asks, 

Is Ben Bernanke the father of Alexandria Ocasio-Cortez?

The reason, 

Mr Bernanke’s unorthodox “cash for trash” scheme, otherwise known as quantitative easing, drove up asset prices and bailed out baby boomers at the profound political cost of pricing out millennials from that most divisive of asset markets, property. This has left the former comfortable, but the latter with a fragile stake in the society they are supposed to build... Soaring asset prices, particularly property prices, drive a wedge between those who depend on wages for their income and those who depend on rents and dividends. This wages versus rents-and-dividends game plays out generationally, because the young tend to be asset-poor and the old and the middle-aged tend to be asset-rich. Unorthodox monetary policy, therefore, penalises the young and subsidises the old. When asset prices rise much faster than wages, the average person falls further behind. Their stake in society weakens. The faster this new asset-fuelled economy grows, the greater the gap between the insiders with a stake and outsiders without. This threatens a social contract based on the notion that the faster the economy grows, the better off everyone becomes... Ten years ago, faced with the real prospect of another Great Depression, Mr Bernanke launched QE to avoid mass default. Implicitly, he was underwriting the wealth of his own generation, the baby boomers. Now the division of that wealth has become a key battleground for the next election with people such as Ms Ocasio-Cortez arguing that very high incomes should be taxed at 70 per cent. 

It is hard to dispute the influence of QE on amplifying these social and political consequences,

According to the Pew Research Center, American millennials (defined as those born between 1981 and 1996) are the only generation in which a majority (57 per cent) hold “mostly/consistently liberal” political views, with a mere 12 per cent holding more conservative beliefs. Fifty-eight per cent of millennials express a clear preference for big government. Seventy-nine per cent of millennials believe immigrants strengthen the US, compared to just 56 per cent of baby boomers. On foreign policy, millennials (77 per cent) are far more likely than boomers (52 per cent) to believe that peace is best ensured by good diplomacy rather than military strength. Sixty-seven per cent want the state to provide universal healthcare, and 57 per cent want higher public spending and the provision of more public services, compared with 43 per cent of baby boomers. Sixty-six per cent of millennials believe that the system unfairly favours powerful interests. One battle ground for the new politics is the urban property market. While average hourly earnings have risen in the US by just 22 per cent over the past 9 years, property prices have surged across US metropolitan areas. Prices have risen by 34 per cent in Boston, 55 per cent in Houston, 67 per cent in Los Angeles and a whopping 96 per cent in San Francisco. The young are locked out... If only the votes of the under-25s were counted in the last UK general election, not a single Conservative would have won a seat.

Watch this space. I have a full book due for release in April, in which the impact of QE is a major part!

Weekend reading links

1. Jean Pisani Ferry strikes a sobering note on the distributional consequences of any green tax deal. Sample this,

The transition to a carbon-neutral economy is bound to make us worse off before it makes us better off, and the most vulnerable segments of society will be hit especially hard. Unless we acknowledge and address this reality, support for greening the economy will remain shallow and it may eventually wane... the distributional effects of the green transition are unfortunately adverse. The poor and the suburban middle class spend more of their income on energy than the rich and the urban professionals do, and often lack the means to buy a new, efficient heating system or to insulate their house. And, because working-class jobs tend to be more carbon-intensive, factory workers and truck drivers will be hurt more than designers and bankers. The problem our societies are facing is massive. It should not be hidden. The French government had to backtrack after the Yellow Vests revolted against a €55 ($63) per ton fuel tax, but a recent estimate of what its needed to decarbonize put the rate at €250 per ton in 2030.

2. The evolution of bigger, saltier and heavier fast food in the US between 1986-2016.

3. The crisis facing large planes like the Airbus A380 is being mirrored in the shipping industry. A twin squeeze, excess supply from a bet on ultra-large ships and declining demand due to slowing trade, threatens the health of global shipping industry.

4. US job market facts,

The U.S. economy has added jobs for 100 consecutive months. Unemployment recently touched its lowest level in 49 years. Workers are so scarce that, in many parts of the country, low-skill jobs are being handed out to pretty much anyone willing to take them—and high-skilled workers are in even shorter supply.

And wages too have been rising

5. This is one of the first full evaluations of the impact of the typical modern fintech on the poor - an evaluation of Kenya's M-Shwari. While there are several assumptions/qualifications (chief being that it covers only a period of 18 months since introduction), broadly the findings are very similar to that from microfinance. 

The short story. On the subsistence or poverty alleviation side, fintech leads to greater access and usage of formal financial channels by the poor and increases their resilience (allows the poor not to have to forego basic expense when faced with various economic shocks). On the consumption and income side, there is no evidence of either increase in consumption nor greater savings and investments in productive assets nor rise in wealth from access to fintech credit. Little impact in terms of its use as working capital or investment in entrepreneurship. 

On its own, the typical fintech innovations (focused on financial inclusion) can perhaps at best be a useful poverty alleviation complement. But its impact in increasing incomes is more likely to be marginal.

6. The landmark Right to Education Act 2009 in India is associated with increased enrolment but declining student test scores. Suggestive evidence on the reasons for the decline in student learning levels - large influx of weaker students or out-of-school students since RTE and the distorted incentives from the introduction of no-fail policy.

7. The Economist has a briefing on the changing landscape of diplomatic, military and economic landscape of Africa.

From 2010 to 2016 more than 320 embassies were opened in Africa, probably the biggest embassy-building boom anywhere, ever. Turkey alone opened 26. Last year India announced it would open 18. Military ties are deepening, too. America and France are lending muscle and technology to the struggle against jihadism in the Sahel. China is now the biggest arms seller to sub-Saharan Africa and has defence-technology ties with 45 countries. Russia has signed 19 military deals with African states since 2014. Oil-rich Arab states are building bases on the Horn of Africa and hiring African mercenaries... As recently as 2006 Africa’s three biggest trading partners were America, China and France, in that order. By 2018 it was China first, India second and America third (France was seventh)... The biggest sources of foreign direct investment are still firms from America, Britain and France, but Chinese ones, including state-backed outfits, are catching up, and investors from India and Singapore are eager to join the fray.

8. The return of industrial policy gathers pace as continental European countries talk of propping up national champions in important industries.

Across the continent, politicians are seeking to influence business using a range of tactics including regulation, nudging managers to do deals and boosting state ownership. At Renault-Nissan, the downfall of Carlos Ghosn has become intertwined with a struggle for control between the French and Japanese governments (see Banyan). Last month Peter Altmaier, Germany’s economy minister, called for champions such as Siemens and Deutsche Bank to be protected. Last week it emerged that the Dutch government has built up a 14% stake in Air France-klm to help its former flag-carrier “perform better”. And Italy is poised to increase to 10% its stake in Telecom Italia, which it began privatising 21 years ago.

Like with most things, there is nothing inherently right or wrong with industrial policy. In fact, like all other policies industrial policy too is necessary and has always played a role. The challenge with industrial policy is getting it right. But that is equally relevant to market capitalist policies like in the US, where regulatory apathy and capture has led to business concentration and uncompetitive markets.

9. Finally, the FT points to the re-emergence of debt problems among frontier markets, this time driven by the expansion of Chinese lending.

Debt levels which had fallen in mid-2000s following restructuring and defaults has now recovered to its earlier alarming levels.

The worst thing about this latest round of debt binge has been, again like earlier times, been channeled into consumption and not investments.

Who said this time is different?

The problem with performance incentives in government

Much has been written about how financial performance incentives can help address state capacity weaknesses and improve outcomes in public systems, especially in certain sectors. 

The mainstream critique to this has revolved around the difficulty of credibly quantifying and then capturing outputs which are both reliable and proximate contributors to the desired outcomes. There is another equally important concern.

Consider this. Public servants, especially field-level functionaries (teachers, medical supervisors, bill collectors, tax inspectors, building inspectors, and so on) are among the highest paid workers (among comparable levels) in any developing country. This also partially explains the scramble for these jobs. In each of these cases, their actual performance is significantly lower that not just their formal responsibilities demand, but much lower than comparators in the private sector. 

So it is commonplace to have a tax collector or inspector assed and realise 30-50% of his estimated demand. Or an electrical engineer manning a substation have distribution losses in the range of 30-40%. Or have a teacher who attends the school only half the days, and whose student learning outcomes are abysmal. Or a primary care hospital doctor achieving just 30% universal immunisation and 40% institutional deliveries. This is business as usual in most low-income countries.

So, you have the highest paid officials achieving such low outcomes, and that too lower than even their much lower paid private sector comparators.  

In this context, economists recommend incentivising these functionaries with performance payments. They proffer evidence from sanitised RCT pilots. These results can appear alluring - logical and neat answers to a complex problem. 

But this is deceptive for at least two reasons. One, its general equilibrium incentive distortion effects. A solution like this immediately shifts the frame of reference from one where the challenge was to improve outcomes using existing levers to one where the challenge is to get the most optimal design of incentives. Implicit in the latter is an acknowledgement that the poor status-quo is inevitable and alright and we are now therefore looking at the performance incentive lever to improve outcomes. The moral hazard from this can create unexpected outcomes. Poor performance soon gets to becoming psychologically rationalised by truant officials. Further, not only the base salary but the structure of incentives too become factors in wage negotiations.

Second, implementation of financial performance incentives at scale can have unexpected effects. The most likely impact is that over time, performance incentives come to be viewed as an entitlement. Or for example, over time, it is reasonable to have expectations and attitudes that rebalance baseline performance downwards and revision upwards of incentives. When an incentive gets scaled up, employees tend to forget the baseline salary (which is then taken for granted) and there is the danger of the incentive becoming seen almost as an entitlement - if you want us to improve the performance, give us more incentives! 

There are several examples of distortions along these lines which have been engendered by financial performance incentives. The practice of bonuses is commonplace in public sector power (generation, transmission, and distribution) companies in India and their impact in terms of actually lowering transmission and distribution losses is perhaps nil. 

When thinking about such neat solutions which we encounter in our daily lives and how they could be transplanted to public sector, we overlook the numerous other formal institutional factors and informal norms that contribute to shaping expectations and attitudes in private sector which in turn contributes to the success of performance incentives there. And we forget that these factors are deficient or absent in public sector, not to mention the fact that public sector is never about selling widgets. 

Update 1 (13.11.2021)

A report by UK's Institute for Government assesses performance bonuses in public systems and finds little evidence of it being effective. 

Nudging on indirect taxes

From The Economist,

Governments around the world are encouraging consumers to ask for receipts by turning them into lottery tickets. Taiwan was an early experimenter, in 1951. The past decade has seen a flurry of such schemes: China, the Czech Republic, Lithuania, Portugal, Romania and Slovakia all now have them. Latvia will launch one later this year. The aim is to make it harder for retail businesses to evade taxes... The problem is not business-to-business transactions; firms can usually reclaim any vat they pay if they keep proper records. But when selling direct to consumers, it is tempting to accept cash without recording the sale... The idea of a receipt-lottery scheme is to give customers an incentive to ask for receipts, thereby forcing sales to be recorded and taxed. Receipts might be printed with a code that can then be submitted into a central draw. Prizes range from decent sums of money to cars and holidays. Digital technology means schemes are cheap to run, even allowing for the cost of prizes... According to a report for the European Commission in 2017, of the ten European countries with the biggest shortfalls in collection of vat in 2014-15, nine have, or are setting up, a receipt-lottery scheme. (Italy is the exception.)

There is little doubt that it will, in most places, have some impact. The degree of impact, and cost-effectiveness, though will depend. For example, complementing this with rigorous data analytics and strong enforcement can potentially have significant impacts.

Capitalism - McKinsey edition!

I have blogged repeatedly highlighting how iconic companies like McKinsey, Goldman Sachs, JP Morgan, Google, and so on have become exemplars of what is wrong with the American capitalism, the variant that others are trying desperately to emulate. Their unfortunate practices, which have amoralised and enfeebled capitalism, may also carry the sparks that, along with other growing trends (widening inequality, financialisation, business concentration etc), could contribute to burning the entire house down. And, in the long sweep of history, that may perhaps be a good thing. 

With McKinsey, from India and Ukraine, and Malaysia and Saudi Arabia, the skeletons have been tumbling out at an alarming pace.

Now comes more news about the conflicts of interest between McKinsey's consulting work and the hedge fund that manages the pensions and investments of about $12.3 bn on behalf of about 30,000 current and former McKinsey's employees, McKinsey Investment Office (MIO).

First it was Valeant, the notoriously price-gouging pharma company, in which MIO was one investor,

Four top Valeant officials, including Mr. Pearson (the CEO), were McKinsey veterans, and the firm was advising Valeant on drug prices and acquisitions. That web of relationships underscores the unusual nature of McKinsey’s hedge fund, and the potential for undisclosed conflicts of interest between the fund’s investments and the advice the firm sells to clients.

The relationship was very deep, almost defining and birthing, albeit indirectly, Valeant itself,

The Senate hearings on Valeant Pharmaceuticals examined how the company, under Mr. Pearson’s leadership, raised the price of Isuprel and Nitropress, decades-old heart medications commonly found on emergency-room crash carts. Mr. Pearson relied on an idea he had honed as head of McKinsey’s pharmaceutical practice: focusing on acquisitions of existing drugmakers instead of spending heavily to discover the next blockbuster drug. Before buying Isuprel and Nitropress from a competitor, Valeant consulted McKinsey, which reported that the drugs had “material pricing potential,” according to records obtained by the Senate. After the acquisition, Valeant raised Isuprel’s price by 720 percent, and Nitropress’s by 310 percent. As McKinsey consultants were advising Valeant in late 2014, two of MIO’s funds, Compass TPM and Compass Offshore TPM, acquired an indirect stake in Valeant, through Visium Asset Management, a fund soon to implode in an insider-trading scandal. Through another outside hedge fund, Aristeia Capital, two other MIO funds acquired shares in Valeant in early 2015, after the company agreed to buy Dendreon, a bankrupt drugmaker. The McKinsey funds were listed among the eight noteholders of Dendreon, entitled to a portion of the $495 million, in cash and Valeant shares, that Valeant paid for the company… Even if the McKinsey fund’s investments in Valeant were indirect, the firm’s partners stood to profit from their own advice. 

Then came the case of Alpha Natural Resources, a Virginia based firm, while filed for bankruptcy. A federal judge reopened the bankruptcy case recently after learning that McKinsey had cheated the court by advising the company without disclosing that through MIO, it was one of the company's secured creditors. And secured creditors had recovered the most most money. It is thought that McKinsey had improperly reaped a profit of $50 million because it helped determine the way secured creditors were compensated. The US Department of Justice (DoJ) has supported the reopening. The federal Judge Kevin R. Huennekens wrote,

“These are some of the most serious allegations that I have ever seen.”  

Another case, one involving Colorado's Westmoreland Coal, too is currently in a Federal Bankruptcy Court in Houston. An activist, Jay Alix, and the US DoJ, are accusing McKinsey of "pervasive disclosure deficiencies" and "incredibly inflammatory" wrong-doing. As the courtroom battle rages, McKinsey has been forced to return $1.2 million in fees to Westmoreland.

The courtroom fight hinges on whether McKinsey or its clients have hidden interests in a bankrupt coal company that the firm has been advising, a practice prohibited by federal laws meant to ensure that one insider can’t effectively cut itself or its friends a great deal at the expense of others... On Dec. 14, the department said in a court filing that McKinsey was fraught with “pervasive disclosure deficiencies” and should be dismissed from the Westmoreland case immediately and stripped of the fees it had earned so far.

Finally, there is the federal bankruptcy case of Puerto Rico, where McKinsey indirectly owns at least $20 m in bonds whose fate is inextricable from the advise the company is giving to the government,

McKinsey & Company is advising the Puerto Rican government on a financial overhaul to lighten its crippling debts — a process that will determine how much money the bankrupt territory’s creditors recoup on their investments. The giant consulting firm has millions of dollars riding on the outcome. The reason: McKinsey owns bonds issued by Puerto Rico. That creates a potential conflict of interest between McKinsey’s client, which wants to save as much money as possible, and McKinsey itself, which wants to make as much money as possible on the bonds... Affiliates of McKinsey revealed their investments in filings in federal court in Puerto Rico’s capital, San Juan. The filings were not made under McKinsey’s name, and were among nearly 165,000 bankruptcy claims. McKinsey so far has received $50 million in fees to advise Puerto Rico’s federal oversight board on drafting a blueprint for the island’s fiscal affairs over the next decade.

Lynn M. LoPucki, a law professor at the University of California, Los Angeles, and the founder of the school’s Bankruptcy Research Database, says,

“It’s a conflict of interest. They are making decisions that will determine how much money is given to themselves.”

The defence is that there are Chinese walls to limit conflicts of interests. But how meaningful are those walls? Sample this,

McKinsey’s main competitors, Bain & Company and Boston Consulting Group, have nothing like MIO. Regulatory filings show that an outside manager, Vanguard, manages their employees’ 401(k) retirement accounts. But investments for McKinsey’s 401(k) plan, which must be disclosed to the Department of Labor, make up only about half of MIO’s assets. The rest, about $6 billion, is the private wealth of McKinsey’s current and former partners... Those partners have access to a trove of inside knowledge about their clients. According to its website, McKinsey counts as clients 90 of the world’s 100 biggest companies. It advises two-thirds of the top mining companies, more than half of the top 25 airlines and 60 percent of the 100 largest banks. At the same time, the current or former McKinsey consultants who dominate MIO’s board include the leaders of the firm’s wealth and asset management and energy practices, according to an S.E.C. filing... an official with the Justice Department’s Office of the United States Trustee argued that the hedge fund was not a “blind trust,” as claimed. The Justice Department pointed out that the head of McKinsey’s bankruptcy practice, Jon Garcia, sat on the fund’s board. Until he stepped down in 2017, Mr. Garcia received regular reports on the fund’s investment decisions and ratified them, the official said.

For a company that promotes corporate best practices, the opacity of MIO’s investments is striking,

For someone who manages such a large and successful portfolio, there is surprisingly little public information about Mr. Todd Tibbetts, the fund’s longtime chief investment officer. So secretive is the fund under Mr. Tibbetts that its managers were discouraged from attending industry events and, even when they did, from talking to participants, one former fund executive said. One reason for the secrecy, another former executive said, is that McKinsey does not want people connecting any dots between what its consultants do and where its hedge fund invests... McKinsey does not disclose the identity of its clients — the chief executives, prime ministers and princes who seek its counsel on management best practices. And even as the firm is privy to market-moving corporate maneuvers and confidential government information, its hedge fund’s investments are often secret, with a large part of its approximately $12.3 billion in holdings concealed behind a tangle of shell companies in an island tax haven in the English Channel. 

Even in this world of utilitarian excesses, this MIO investment structure in a dubious Chinese entity, in particular, is very hard to rationalise,

And despite all this, from just the past six months, sample this for cheekiness,

McKinsey says the way the fund is structured and operated ensures that its employee investors do not stand to benefit from the firm’s inside knowledge and consulting advice. Hedge fund managers do not coordinate with McKinsey consultants, the firm says, and about 90 percent of MIO’s capital is managed by outside funds. “MIO and McKinsey employ separate staffs. MIO staff have no nonpublic knowledge of McKinsey clients,” the company said in a statement. “For the vast majority of assets under management, decisions about specific investments are made by third-party managers.”

Actually the Times investigation is extremely generous in so far as it does not go the full distance with interpreting a whole series of potentially criminally indicting corroborative circumstances. For example, it stops short of highlighting the coincidence between McKinsey's consulting role and MIO's investments in these relatively obscure four contexts. Given that MIO's portfolio is $12.3 bn and not $12.3 trillion, it is intriguing as to why should MIO ever even have considered investing in these odd assets? It is difficult to believe that when the MIO fund managers scanned the global investment opportunities landscape, these four obscure assets would have stood up as standout investment opportunities. The coincidence raises the question of whether MIO got the consulting business for McKinsey or did the consulting business open up the investment opportunity? Either ways, it is a striking illustration of the corrosive business practices that exemplars of capitalism follow. 

One is perhaps an inadvertent omission, to be condoned. Two, three, and four (and more waiting to tumble) is perhaps more than enough to attribute a conscious pattern for criminal indictment.

But such is the regulatory and political capture that it is futile to even expect heads to roll internally, much less criminal prosecution. Most likely those partners and executives would get even bigger bonuses and promotions. 

Update 1 (29.12.2020)

Daniel Markowitz has an article in The Atlantic which examines how McKinsey's technocratic management work undermined the society and the middle class. 

Fundamentally, management consulting, with efficiency maximisation being the preeminent consideration, elevates one particular value or attribute, at the cost of many others like fairness, empathy, loyalty, and so on. 

Weekend reading links

1. No surprises that the plumbing of Indian capital markets is of dubious quality. Sample this about credit ratings of Indian corporates,

While India has 70-odd companies that are rated highest quality, only two companies in the US enjoy this distinction. No company in Germany and UK enjoys AAA rating. Among emerging countries, China has only 14 AAA-rated entities. This implies a gulf between credit standards in India and elsewhere. The exacting standards observed in other countries are missing among domestic agencies.

2. Economist on African countries building up cash transfers based social protection programs,

From 2010 to 2015 the countries of sub-Saharan Africa launched an average of 14 schemes per year, up from seven per year between 2001 and 2009. These countries spend an average of 1.2% of GDP per year on social safety-nets, using a broad definition that includes pensions as well as support for children and the poor. That is only a little less than the average for developing countries (1.6%).

In countries where state capacity is chronically weak and public provisioning of services of appalling quality, cash transfers may not be a bad idea. And in any case, as a social safety net, it is a good idea. The only question is whether these countries have the fiscal space to provide this as well as the regular public goods.

3. Tax avoidance, Amazon edition,

The company, with nearly $11 bn in earnings in 2018 paid nil corporate tax. Unlike others who transfer profits to off-shore subsidiaries or undertake reverse mergers, Amazon takes advantage of Section 162 (m) of US tax code by expensing stock-based compensation.

4. The Adani Enterprises has bagged the 50 year concession for operating all the 5 currently AAI operated airports - Lucknow, Jaipur, Ahmedabad, Trivandrum, and Mangalore - which were put up for auction.  They quoted the highest per-passenger fee for all the five airports, bidding "aggressively" to bag the contracts in a very competitive race. The tender also includes rights for developing real estate projects and operating retail within the airports. Adani Enterprises, which has no experience of running commercial airports, also becomes the largest airport operator in India.

I am not sure this is a good turn of events for multiple reasons. Is the bid motivated by the secondary activity of real estate and commercial development of the airport assets or development of the airports themselves? Does the Adani group have the ability to develop the financial strength over the next 3-4 years to ensure that is effectively delivers on its short to medium-term contractual obligations (almost Rs 12,500 Cr investments in next five years), given their already big commitments on new ports, gas distribution, power transmission, and now airports, not to speak of its already over-leveraged (Rs 1.2 trillion group debt) legacy businesses? Finally, while perhaps too early, are competition authorities not concerned by the Group's dominance (or likely dominance) in several sectors?

5. On premature de-industrialisation,

Just 43 per cent of trade in textiles and apparel was based on labour-cost arbitrage in 2017 (defined as exports from countries whose GDP per capita is a fifth or less than that of the importing country), compared with 55 per cent in 2009, as the first chart shows. For furniture, toys and other labour-intensive goods, there has been a decline from 43 per cent to 35 per cent over the same period.... McKinsey calculated that trade intensity (the ratio of gross exports to gross input) rose rapidly from 1995 to 2007, when it peaked at 28.1 per cent. It has since fallen back to 22.5 per cent. In addition to this, trade has become more regional. The share of goods traded within the same region fell from 51 per cent in 2000 to 45 per cent in 2012, MGI said, but has rebounded to 47.7 per cent, led by Asia and Europe. This reduces the scope for emerging countries to sell to the rich west.

6. The International Court of Justice rules 13-1, in a non-binding judgement, that Britain's occupation of its 'last colony in Africa', the Chagos Islands in the Indian Ocean, is illegal. As FT writes,

In a non-binding opinion delivered in The Hague on Monday, the court said the UK’s half-a-century of rule over the islands, including expelling 1,500 inhabitants and hosting a US military base, was a “wrongful act” and not “based on a free and genuine expression of the people concerned”. “The UK is under an obligation to bring an end to its administration of the Chagos Archipelago as rapidly as possible"... Although advisory, the ruling will increase pressure on Britain to cede the islands to Mauritius, from which it split the territory... In what the court on Monday called an “unlawful detachment,” London split the Chagos Archipelago from Mauritius in 1965, three years before the end of colonial rule. From 1968 to 1971 UK authorities forced Chagossians to leave — gassing their pets — to make way for a US base on the largest island, Diego Garcia. At the time a British diplomat dismissed the islanders as “some few Tarzans or Man Fridays whose origins are obscure and who are hopefully wished on to Mauritius”.

7. David Leonhardt examines the trends associated with incomes of different income segments of the US population.

The incomes of the upper middle class have grown at the same rate as per capita income growth. His recommendation,

Politicians should recognize that there are three broad income groups, not just two. The bottom 90 percent of Americans does deserve a tax cut, to lift its stagnant incomes. The top 1 percent deserves a substantial tax increase. The upper middle class deserves neither. Its taxes should remain roughly constant, just as its share of economic output has.

8. MSCI announces an increase in China's weighting in its flagship $1.9 trillion EM index from 0.71% to 3.3% by November 2019, a decision that is expected to see an estimated $125 bn of offshore funds flowing into Chinese equities this year. 

9. Finally, a fascinating Simon Kuper peek inside the great Barcelona football club and how, through its Innovation Hub, the club is creating not just the "future of football" but an entire innovation eco-system focused on sports. Sample this,

Barça field men’s, women’s and children’s teams in sports from basketball to roller hockey. The start-ups or universities with which Barcelona are partnered can test their findings on athletes. Sometimes Barça’s staff help co-develop a product. If any of this work results in a breakthrough in, say, treating hamstring injuries, then Barça’s athletes will benefit first. After that, though, the club hope to spread any new products throughout global sport... If a company can market a product as “tested at FC Barcelona”, the club will charge royalties. And if the product is in sleep or nutrition, areas in which elite athletes and ordinary mortals have similar needs, revenues could be large. Now Barça plan to launch investment funds, with initial outside capital of €125m, to invest in tech and sports projects worldwide.

In terms of a match objective, this is stunning,

The team’s basic playing style barely changes from match to match. They aim to play a passing game in the opponent’s half and to lose the ball in dangerous positions no more than six times a match, allowing the opposition about three threatening attacks. More than six, and Barça aren’t playing well.

The remarkable thing is how much value is being attached to data analytics, at least for now, in the club's successes. In fact, given this, such humility of coach Ernesto Valverde stands out,

“This is a game where the coach has less margin than any coach in another sport, because I’m shouting to the player over there, and he doesn’t hear me, and the one by my side doesn’t either. This is a continuous sport in which the coach has barely any influence, or at least much less than in basketball: we only have three substitutions, the game never stops [for time-outs]. So, football belongs to the players. For 45 minutes at a time, non-stop, the player takes his own decisions. I have to say that the great players analyse the game better than I do.”

Sample this on the limitations of data analytics,

Attacking in football is about creating superiorities. These can be numerical (two of your players against one of theirs), positional (your player controls a space) or qualitative (Messi dribbling against an inferior opponent). Barça hope that tracking data can uncover ways to create superiorities. For now, the club’s analysts can barely help the players do that. On the contrary: the analysts learn about football by observing the most intelligent players. For instance, Barça’s midfielder Sergio Busquets, the team’s pivote, knows just how to draw an opponent towards him and then release a teammate into the space the man has left. No coach shouting instructions at him through an imaginary earpiece could advise him better. How to identify intelligent footballers? A Barcelona official fantasises about one day tracking players’ brains. For now, though, the analysts suspect that the most intelligent players — Busquets, Andrés Iniesta or Messi — are those who almost always face the right way on the field.

This is brilliant sports writing. Do read the whole article.

The future of mobility - automotive industry to mobility industry?

McKinsey has an article on the second inflection in mobility, a full century after the first shift from horses to internal combustion engines,

Electric and autonomous vehicles, more interconnected and intelligent road networks, new customer interfaces and services, and a dramatically different competitive landscape in which tech giants, start-ups, and OEMs mix and mingle are just a few of the shifts in store. Radical improvements in cost-effectiveness, convenience, experience, safety, and environmental impact will, taken together, disrupt myriad business models on an almost inconceivable scale.

How will this prophecy stand the test of time?

For precedent, sample this about the spectacular growth of motor vehicles, on the back of Henry Ford's Model T,

In 1900, about 4,000 automobiles were produced in the United States; none of them were trucks. During the 1910s, the number of automobiles across different parts of the United States began to eclipse the number of horses and buggies. By 1920, America had more than 9.2 million registered motor vehicles, including more than one million trucks.

And its social, cultural, and economic impact,

On the left side of the growth curve were steam and hay (and feet), dirt roads, manure-filled cities, and sleepy countryside. Past the inflection came gasoline, paved roads and highways, motels, fast-food restaurants, and suburbia. Mobility was not just cars, but parts manufacturers and suppliers, mechanics, taxis, buses, commuter railways, and, in time, metro-area airports. The auto industry created millions of jobs and massive new profit pools. Three of the top ten highest incomes reported to the Internal Revenue Service in 1924 came from automobile-industry titans. 

So what is the promised world of mobility industry,

Much of the activity revolves around automotive technologies known by the acronym ACES—vehicles that are autonomous, connected, electric, and shared... Autonomous-car technologies will soon transform what “riding” means... The technology for connectivity—what car riders experience along the way—is also poised for a breakthrough... Occupants will be afforded personalized infotainment through voice and hand gestures... connectivity systems will become a “virtual chauffeur,” in which cognitive artificial intelligence (AI) can anticipate and fulfill riders’ needs... Sensors embedded in vehicles will communicate with traffic lights, street signs, and each other, allowing cars to travel much closer together and shortening travel times significantly. As weather conditions change and car volumes ebb and surge, routes will be optimized near instantaneously... Significant improvements in battery technologies and the use of renewables, plus evident regulatory will from many governments to impose regional and global carbon limits, means the likely end of ICE’s technology predominance. That spells new opportunities for those in the energy and metals and mining industries, among others... If ride-sharing AVs eventually remove the driver from the equation, they could make riding more affordable... That will greatly lower the total cost of ownership due to lower variable cost, despite higher fixed cost (the original price of the car)... More people e-hailing means one less major use case for OEMs to address, which will allow affordable, purpose-built vehicles to be introduced in turn. Shifting from “pay for vehicle” to “pay per mile” has the happy result of reducing the price of both.

Now to wait and see which of these trends emerge!