Sunday, September 25, 2016

Weekend reading links

1. Elizabeth Warren's tour de force examination of Wells Fargo Chairman John Stumpf's accountability in the cross-selling scandal, which created more than two million bank and credit accounts without customers consent, is a must watch! It is a great example of well-reasoned and well-researched polemic.
Unfortunately, Ms Warren's scathing attack and having to squirm through it may be the only punishment that Stumpf may have to endure. 

2. Zero Hedge points to this incisive presentation by Stanley Druckenmiller, where he charts the excesses that have been building up the US economy. This representation of the skewed nature of the corporate resource allocation during the ongoing business cycle says it all - $2 trillions spent on buybacks and acquisitions in US in 2015 against $1.8 trillion on R&D and office equipment!
The irony is the equity market boom despite all the adverse headwinds,
The corporate sector today is stuck in a vicious cycle of earnings management, questionable allocation of capital, low productivity, declining margins, and growing indebtedness. And we are paying 18X for the asset class.
3. Mohamed El Erian points to this graphic about IMF projections since the global financial crisis which captures the growing pall of gloom surrounding the world economy.
4. George Perkovich has a summary of the options available for India following Pakistan's latest foray,
India could... consider limited air and special forces strikes on known terrorist facilities and Pakistani military installations in the part of Kashmir that Pakistan controls. The campaign would surely produce television images of destruction that could satisfy (temporarily) the Indian desire for revenge. But a restrained use of force could signal lack of Indian resolve, thereby emboldening the Pakistani military and providing it with little incentive to crack down on anti-Indian groups. More daringly, Modi could send planes through Pakistani air defenses to bomb militant groups’ facilities in the heartland of Punjab—assuming that India had credible evidence to link Sunday’s attack to those groups. But such an action would likely prompt Pakistan to mount counterattacks on India, again risking escalation whose potential destructiveness would be out of proportion to the injury India has suffered. 
India could, and probably will, increase the intensity of covert operations to foment disorder in Pakistan, particularly in the restive province of Balochistan. Such activities would certainly harm the interests of the Pakistani military. But they would also bolster Pakistan’s effort to portray India as morally and politically equivalent to Pakistan in the use of terrorism, a label India has long sought to avoid. India will also justifiably seek to mobilize the world against Pakistan as a state-sponsor of terrorism, which is increasingly difficult to deny. But the long history of Kashmiri resistance to Indian rule and the fact that the target of this weekend’s attack was a military installation—not civilians—complicates the Indian narrative.
The last one is clearly the most prudent option, though it would require escalating the rhetoric against Pakistan at international forums to higher level. The flip side to this would be that India would be spending its scarce diplomatic capital combating Pakistan, at the cost of other more strategic issues. Exactly what China would want - leave India's diplomatic energies expended fighting Pakistan. 

One way to manage the challenge is to respond asymmetrically - leave the junior-most Indian diplomat stationed with the UN Mission respond aggressively to the Pakistani Prime Minister, no less. This was brilliant. Unfortunately, there are limits to such opportunities. 

Staying on the same issue, the mainstream western narrative of "Kashimiri resistance" to "repression by Indian security forces" that the likes of Perkovich peddle is as much a trivialisation of the debate as the populist incendiary rhetoric in mainstream Indian media. After all, Kashmir since independence has always been ruled by Kashmiris. 

5. Andy Mukherjee points to the maladies of corporate governance in India and how the equity markets have become the quiet platform for its most egregious excesses. The case in point is the merger of Max Life Insurance and HDFC Standard Life to create India's second largest insurer, albeit at just 6.75% of market share. The fine print of the merger involved a $127 million payout from the merged entity as non-compete fee to Mr Analjit Singh, the promoter of Max Life Insurance. This is despite Mr Singh continuing to own a 6.5% stake in the merged entity. 
Who cares if Singh started India's 25th life insurance company? As many as 14 of the existing two dozen players control less than 1 percent of the market each, while Life Insurance Corp. of India -- the former state-owned monopoly -- still collects 70 percent of all premiums. With a 6.75 percent share, the bulked-up HDFC Standard Life would have nothing to worry about if Singh did decide to start a rival. He would be crazy to want to. Besides, it isn't just Singh and his family who are being compensated. Even his private investment vehicles, such as Mohair Investment & Trading, will collect a part of the $127 million, leaving little doubt about the true nature of this side payment. Had this been a regular takeover of the business by HDFC Standard Life, the stock-market regulator would have said no to the sham noncompete. But since the deal has been designed as a merger, it's up to the other shareholders to try to block the payout.
6. Fantastic graphical presentation of the spectacular improvements in global health indicators over the past two centuries. This one on the improvements in life expectancy is truly impressive.
7. Finally, one more from outstanding Twitter handle of Ian Bremmer, this time an illustration which captures the staggering complexity of power dynamics in the Middle East. Maybe Israel should relax!

Tuesday, September 20, 2016

The distance between financial access and usage

More evidence from Chile, Malawi, and Uganda that merely giving bank accounts does not translate automatically to its usage, leave alone outcomes like increased savings,
Bank accounts as currently offered appear unappealing to the majority of individuals in our three samples of unbanked, rural households – even when these accounts are completely subsidized. While we do observe substantial usage among a subset of active users, we are unable to pick up any statistically significant effects on downstream outcomes. This is not surprising since the average impact on total savings is itself relatively modest at best, and noisily estimated. If any, treatment effects on downstream outcomes are likely diffuse since savings purposes are heterogeneous, thus difficult to detect, because of little ability to predict how user households would use the savings.
The authors speculate about two possible reasons - poverty (people are too poor to save any money and thereby use the accounts) and transaction costs (bank branches are far away and the processes associated too complicated).

I would go one step further and add that even if we assume away these two, there is another barrier, which we haven't even seriously though about - demand. Do we have financial products that are tailored to meet the requirements of poor people, especially those reliant on farm incomes? Forget product development, there is very little serious thinking on this last mile gap. 

Fundamentally, the solution to a problem has to start with the product (or service) that caters directly to the issue. In this case, poor people need financial products that meet their requirements, and not just processes and intermediate steps that lead to the desired products. I have blogged earlier about exactly such interventions that meet their requirements.

So, the high prevalence of unused accounts from India's Jan Dhan Yojana, even in areas where poverty is not as acute and where access and transaction costs are marginal, should not come as any surprise. We under-estimate the formidable last mile gaps that come in the way of addressing such problems. 

Sunday, September 18, 2016

More on the excesses of transparency

Sometime back I had blogged urging caution on excessive accountability and transparency. Ajay Shah draws attention to an excellent paper by Cass Sunstein where he draws the distinction between transparency in the inputs and outputs of government activities and draws up the costs-benefits calculus,
Government can be transparent about its “outputs”: its regulations and its policies, its findings about air and water quality, its analysis of costs and benefits, its assessment of the risks associated with cigarette smoking, distracted driving, infectious diseases, and silica in the workplace. It can also be transparent about its “inputs”: about who, within government, said what to whom, and when, and why. The argument for output transparency is often very strong, because members of the public can receive information that can help them in their daily lives, and because output transparency can improve the performance of both public and private institutions. Where the public stands to benefit, government should be disclosing outputs even without a formal request under the Freedom of Information Act. In fact it should be doing that far more than it now does. The argument for input transparency is different and often weaker, because the benefits of disclosure can be low and the costs can be high. There is good reason for a large increase in output transparency -- and for caution about input transparency.
Sunstein quotes James Madison to make the point very powerfully,
It was . . . best for the convention for forming the Constitution to sit with closed doors, because opinions were so various and at first so crude that it was necessary they should be long debated before any uniform system of opinion could be formed. Meantime the minds of the members were changing, and much was to be gained by a yielding and accommodating spirit. Had the members committed themselves publicly at first, they would have afterwards supposed consistency required them to maintain their ground, whereas by secret discussion no man felt himself obliged to retain his opinions any longer than he was satisfied of their propriety and truth, and was open to the force of argument. . .. No Constitution would ever have been adopted by the convention if the debates had been public.
More fundamentally, information about inputs cannot be abstracted in isolation from its transactional context, whose literal reproduction is virtually impossible. In the absence, a specific input is vulnerable to being grossly misinterpreted and detract attention from the issue under consideration. This risk becomes amplified in a world of high voltage media spins and trials.

Consider the example of the deliberations of an internal committee of officials who have been appointed by the Government of India to give recommendations to fix an off-set price for the oil and gas exploration licenses. In the prevailing environment where the popular norms in favour of revenues maximisation and not benefiting private companies are entrenched, officials will be apprehensive of being marked out in public as having taken an opposing view and be insinuatingly referred to having favoured large private companies (in this case, it could be one particular large private group!).

One such view could be the legitimate opinion that an oil importing country like India should instead of maximising revenues seek to increase domestic production and thereby reduce import dependency, even if it runs the risk of benefiting the private explorer if it gets lucky and is able to strike resources at low extraction costs. This view could be reinforced by recent oil market developments - weakness in the oil prices, and failures to attract interest in much larger off-shore block auctions in Brazil and Mexico. I, for one, am strongly inclined to this view, as I have blogged earlier here.

Or consider a decision involving adjudicating on the imposition of a standard on effluent discharge from certain industries. One side, popular among environmentalists, favour zero liquid discharge (everything is reused) for certain categories of industries, which has become entrenched in popular media as the socially preferred option and anything else as a concession to polluting and corrupt firms. This has also, for whatever reasons, assumed significance in light of an affidavit submitted by the Ministry of Environment and Forests to the Supreme Court of India. However, practical considerations and experience from elsewhere would show that this is prohibitively expensive and may not be a prudent choice. But supporters of this view face the possibility of being branded as corporate stooges and vilified as corrupt. 

More fundamentally, any public policy decision involves an exercise of weighing of public costs and benefits, at least some of which are realised over long durations and also involve significant private benefits. Complicating matters further, it is virtually impossible to make estimations with great accuracy and often times the realisations are way off from the original estimates. In such a complex decision making environment, it is important that decisions get taken after examining all dimensions and sides involved through a most open and liberal deliberative process. It is inevitable in such deliberations that people take positions and make arguments which would evolve, even change diametrically, along the process.

Making all this public, with the attendant risk of insinuating extracts quoted off context finding its way into a public media debate, would invariably affect the openness of the deliberative process by making officials reluctant to be on record with such views. For example, in the case of the debate on effluent discharge, at the least, the sceptics of the zero discharge would have played the critical role of steering the decision somewhere to the middle of the spectrum. They would have helped avoid populist extremism in the choice of effluent standards, and balance the requirements of environmental protection and economic growth.

It is for this reason that the US Freedom of Information Act provides for exempting privileged communications within or between agencies, including deliberative process privilege, from its jurisdiction. India's Right to Information Act however does not have this exemption. And its effects have been not to benign. Given that the 10th anniversary of the RTI Act just got over, it may be an opportune moment to revisit its provisions and refine them. Unfortunately, the political courage required for this may not be very forthcoming. 

Thursday, September 15, 2016

The recalibration of the pensions privatisation debate

It is a natural tendency of human beings, especially as a social collective, to seek out simple solutions to complex public policy problems. What if there are no simple solutions to complex public policy initiatives?

Consider the example of pension schemes. The conventional wisdom since the early eighties has been that pension plans are best administered by private fund managers. As part of the Reagan-Thatcher axis and later the Washington Consensus, countries were encouraged to move from defined benefit to defined contribution schemes as well as privatise the management of their pension funds. Chile was cited as the poster child of such reforms.

Now, three decades later, Chile is revisiting its pension model. The Times has a nice article,
The options being considered include creating a state-run pension administrator, raising the retirement age, instituting a 5 percent contribution from employers and adopting stronger regulations for the pension fund administrators. “After going from a totally public system to the other extreme in 1981, now we are moving towards a middle ground that combines individual savings, state spending and contributions from employers,” said Mr. Bravo, of the Catholic University of Chile. “Another option is tearing down the A.F.P. system, but it’s too costly. We don’t have the privilege of starting from zero anymore.”
Clearly, the privatised pension model overlooked the messy realities of managing a system with too many moving parts, which cannot always be either regulated or incentivised into conformity. It also mistook a certain demographic and global development moment to be representative of all times to come. 

A pension account requires contributions to be made into individual accounts with a defined periodicity. It is no surprise that poor people, suffering from real financial constraints and amplified behavioural biases, struggle to meet this requirement. The example of unutilized bank accounts from India is an example of such serious last mile gaps. And now, there is growing evidence from everywhere that employers, more due to fiscal and unscrupulous reasons, too renege on contributions.  The article mentions that in Chile, a study found that only a quarter of those who retired last year had paid into the pension scheme for more than 25 years and 62% of women contributed for less than 15 years. So, it appears that the private providers are not well positioned to bear these risk all by themselves. 

Then there are structural issues. The demographic pyramid has slowly changed shape, leaving an increasing share of subscribers exiting the labor market. The age of double-digit or high single digit returns on portfolios looks most likely a thing of the past as we peer into a long period of low interest rates. In fact, as the management and other fees extracted by fund managers rise as a proportion of net returns, pension funds are left clutching at the straws. This has naturally re-iginted the debate on whether the contributions of the workers are in the first place adequate enough to sustain a pension scheme which can give reasonable payouts at the time of retirement. 

All this naturally questions the priors about the neat outsourcing of pensions management responsibilities to private providers. At a more fundamental level, this should be one more in the long list of reminders that the quest for simple, neat, outsourced, start-up driven solutions to complex problems is most likely to be futile.

Update 1 (18.09.2016)

Times has a nice story on the calculation of pension liabilities. The actuarial measurement, which uses a discount rate of 7-8 per cent (the expected investment return), vastly understates the actual liabilities. In contrast, if the risk free rate is taken as the discount rate, the net present value of the liabilities would have to be discounted by closer to 2-3 per cent, which would in turn be a much larger liability. Pension funds prefer to state the former and keep the latter confidential.

Wednesday, September 14, 2016

Mid-week reading list

Back to blogging by catching up on news.

1. So, finally it happened. Henkel and Sanofi became the first two non-government companies to sell corporate bonds at negative rates. John Mauldin has a nice reminder about the damage being inflicted by NIRP.  

There is a first order dissonance. The assumption that extended period of low rates will help buy time to repair balance sheets, boost consumption, and thereby stoke an investment cycle may be way off the mark. In its absence, low rates are merely pushing at a string, engendering distortionary resource misallocations everywhere. This has no pleasant end-game. Strong words from Ananth.

2. Martin Wolf, in a disappointing column, has this graph from the work of Lukas Rachel and Thomas Smith of Bank of England on the different contributors to the near five percentage points decline in global neutral interest rates since 1980. 
This graphic, as Wolf likes to point out, shows that there is a secular decline in real interest rates. But an analysis of the contributing factors should also highlight the futility of trying to stoke demand and investment by keeping rates low.     

3. Wise words from Larry Summers,
There is a case for experimenting with mobilising private capital for use on infrastructure that has been a public-sector preserve, such as airports and roads. But, the reality that government borrowing costs are much lower than the returns demanded by private-sector infrastructure investors should lead to caution. It would be unfortunate if, in an effort to avoid deficits, large subsidies were given to private financial operators. Only when private-sector performance in building and operating infrastructure is likely to be better than what the public sector can do is there a compelling argument for privatisation.
Though made in a different context, the underlying message that cautions against the indiscriminate advocacy of public private partnerships and private investments in infrastructure. As we wrote sometime back, PPPs make sense only when there are real efficiency gains to be had. Unfortunately, governments have come to rely on PPPs as a means to overcome their fiscal constraints.

4. Interesting twist to the Milanovic "elephant graph" that had become the focus of debates on widening inequality and the role of globalisation. A new study by the Resolution Foundation appears to show that the hump is largely the result of China's spectacular rise and the trough is due to the declining or stagnant incomes in post-Cold War Eastern Europe and Japan. Stripped of these, incomes appear not to have done as bad.
This revision does not upend the globalisation-has-produced-winners-and-losers story. There are too many smaller stories within countries, which are likely to convey a more accurate story. 

5. Amidst all the talk of the "fastest growing large economy", the underlying reality for India may not be very healthy. Ananth points to this dismal trend on corporate investment from the latest RBI bulletin.
Private investment is unlikely to rebound anytime soon given the twin balance sheet problem that has ravaged banks and corporates. While the non performing asset problem of banks is well-known, what is less known is the country's corporate sector woes. The FT writes,
According to Ashish Gupta, Credit Suisse’s Mumbai-based financial analyst, almost one-third of Indian companies are “chronically stressed” — defined as lacking the cash to cover interest payments in four of the past eight quarters. Most of these are infrastructure companies that over-borrowed on the erroneous assumption that India’s economy would grow quickly, sustaining demand for the roads and power they sought to provide. They did not envisage unhelpful government policies and payment delays. Today, their plight, and that of their lenders, is preventing the flow of credit needed to fuel further economic growth. As a result, the entire country has become caught in a lending gridlock, since few banks have the capital to absorb a huge writedown of their bad loans.
The article points to HSBC having NPAs cross ten percentage points, levels already breached by many public sector banks some quarters back. Scratch the surface more vigorously, and do not be surprised if some of the largest private sector banks join the group.

6. Nice weekend article in FT which captures the age of Rodrigo Duterte. Boris Johnson, the British Foreign Secretary, has interesting things to say about the two US Presidential candidates. On Donald Trump,
“The only reason I wouldn’t visit some parts of New York is the real risk of meeting Donald Trump.”
And on his rival, Hillary Clinton, 
“a sadistic nurse in a mental hospital”.
7. Finally, Tim Harford points to a new term for the 'economics of rottenness', kakonomics. It is a state where economics get caught in a low level equilibrium of mediocrity, a kakocracy! Could extend this term to all social science.

Monday, September 5, 2016

EM Sovereign Bond Issuance

FT captures the latest round of capital flight into emerging economies, 
A punishingly low yield environment for money managers has sparked a jump in demand for emerging market fixed debt in the past few months, as lack of inflation keeps interest rates in big economies on hold and prompts additional monetary easing from the European Central Bank, the Bank of Japan and the Bank of England. In the US, weak jobs data published on Friday have also pushed back expectations of the Federal Reserve raising interest rates this month, potentially removing a threat to the buoyancy of emerging market assets. Record inflows of funds have in turn pushed up emerging market bond prices, reducing borrowing costs down and spurring an increase in debt sales.
So far this year, they have issued bonds worth $90 bn, and it is estimated to top a record $125 bn in hard currencies like dollar and euro by end of the year.
The major drivers are Latin America fueled by Argentina's return and Middle East and Africa led by Saudi Arabia. These are also the areas most vulnerable to economic troubles when the capital flows cycle reverses, as it will. In contrast, debt issuance by emerging Asia and Europe has been declining. 

It highlights an interesting challenge. On the one hand, the most prudent EM economies exercise great restraint in accumulating debt. But on the other hand, at any given opportunity, the usual suspects leverage up excessively. The net result is that EM debt exposures of large funds rises during such episodes of capital flows into emerging markets, leaving EM assets as a category vulnerable to capital flow reversals. Contagion effects are inevitable on even the fiscally prudent economies. 

The fiscal prudence of governments in these economies, however, masks the rapid accumulation of external debt among non-financial corporates. It is this channel that becomes the primary source of spillovers as the sudden stops emerge. 

Sunday, September 4, 2016

Engagement opportunities in education

Which areas in school education can deliver the biggest bang for the buck? For non-governments seeking 'big bang' engagements in education, I can think of three exciting opportunities.

1. There is now a well endowed body of research which confirms that children learn along varying trajectories and speeds. This means that the standard one-size-fits-all classroom instruction is very inefficient. Instead instruction should be customized to the specific need of each child. But the human resources necessary to do this makes it a non-starter, especially in developing countries with large classroom sizes and vast differences in learning levels. It is in this context that adaptive learning systems assume significance.

I am going to stick out my neck and say that adaptive learning systems are the future of education, at all levels. In simple terms, an adaptive learning system gradually leads the student up his or her unique learning trajectory - each student has a specific learning trajectory which is dependent on antecedent learning levels and the student's pace of learning - using the computer as the interactive learning device, with appropriate facilitation by the class teacher.

The effectiveness of any adaptive learning system depends on its learning trajectory algorithms, which is in turn dependent on the volume and diversity of student learning data from which it was developed. As the depth and breadth of data analysed increases, more variations of concept learning mistakes and their speeds of learning becomes available, which in turn enhances the library of learning trajectory algorithms. At its most efficient and ideal, the library should be rich enough that a student can start a grade and move on the most appropriate learning trajectory to achieve all competencies for that grade in the shortest possible time.

2. Fundamentally, achieving grade-specific basic competency in math and languages is about mastering a few concepts. A concept can be anything like place value in decimals or carry forward addition, and there are likely to be no more than a few such ladder concepts in each grade. Children generally make a few standard mistakes in each concept. The challenge for a teacher is to identify the handful of commonest mistakes made in a subject by children in each grade and then teach them in the most effective manner.

In the circumstances, a repository of all such ladder concepts for each grade and their respective common mistakes, with the most powerful teaching methodologies and digital content assume significance. If this can be delivered in an easily accessible (say, on mobile phones) and cognitively striking manner as byte-sized packets (2-3 minute interactive graphic videos) to instructors, it can become powerful instruments to help effectively explain difficult concepts to children.

There are a number of technology entrepreneurs who claim to have apps which can deliver such content. I am not sure whether they have got it right or can get it right so easily. For a start, even identifying the concepts and the commonest mistakes requires access to massive volumes of data. Then there is the issue of developing the most effective teaching methodologies that incorporates learnings from the pathway of each mistake. This content then has to be rendered in a cognitively striking manner. Finally, the content should be delivered in a manner that is easy for teachers to access and encourages them to use.

Not that easy for technology entrepreneurs, howsoever smart, to develop for a few million in a few months. If some of them are still attracting interest, it is only because the market is so large and broad, and largely un-served, that content with even limited value addition will attract significant interest. 

3. Public school systems collect massive volumes of data about various school level activities, including on physical infrastructure and human resources as well as on transactions like student and teacher attendance and utilization of allocation funds. Unfortunately, in the absence of any work-flow application, this data is rarely used as decision-support on either regular monitoring or policy design. I have blogged earlier about the challenges with developing IT applications in education, health, and other sectors.

Given that the monitoring hygiene factors are largely identical across school systems, apart from being powerful force multipliers in improving service delivery, IT applications that incorporates these activities can break open a very large market. But it will have to emerge through a patient iterative process that marries both professional expertise and programming skills with strongly engaged public stakeholders.