The coming pensions crisis

The decades long secular decline in real interest rates has been taking a toll on savers. The pay-as-you-go, defined benefit pension schemes of many local governments, other public entities, and older corporations have promised very high returns whose realization may be difficult in this environment. Hyper-aggressive wealth and asset managers have massive funds under management that promise long periods of high returns, even after excluding their exorbitant management fees. 

The business models in pension funds, insurance, and asset management look likely to be upended by the persistent and declining trend in interest rates. 

A McKinsey Global Institute report last year made alarming prognosis about investment returns in developed markets over the next two decades when compared to the past thirty years. 

Accordingly, the MGI report estimates that for US and Western Europe, annual return on equities and fixed-income securities could be respectively 150-400 and 300-500 basis points lower. Its implications include,

A two- percentage-point difference in average annual returns over an extended period would mean that a 30-year-old today would have to work seven years longer or almost double her savings to live as well in retirement.

And in the medium term and even including emerging markets, as the seven-year return predictions of Jeremy Grantham shows, the likely real returns are mostly in the negative territory.

The biggest victims of this low return environment are likely to be pension funds, especially the defined benefits plans offered by government agencies and legacy plans of the big private corporations. A world with low yields for the foreseeable future compounds the problems for pension funds across the world. In recent years, the dynamics of demographics – rising share of aging populations – has already been exerting pressure on these pension funds, many of which are grappling with massive underfunded liabilities. They now have to earn sufficient returns not only to prevent the funding gap from widening further , but also make incremental returns to bridge their deficits. But the low return environment leaves them without enough to meet even the first objective. Further, the older towns and cities have been facing the problem of stagnant or declining populations which put pressure on their tax and other revenue sources.  

A study published by AQR Capital Management finds that defined contribution (DC) pensioners have to double their current savings to achieve the same target retirement income replacement ratio (RR) of 75%. The historical real return of 7.5% from equities is expected to decline to 5%, and that from bonds from 2.5% to 1%, thereby generating a two percentage points lower annual return of 3.5% on a standard asset allocation of 60% US equities and 40% Treasury Bonds. This means that the worker now has to save 15% a year, not the current 8%, to reach the same RR. On a 2.5% real return assumption, the contributions would have to rise to 19% overall!  

California Public Employees’ Retirement System (CalPERS), the biggest US public pension fund with nearly $300 billion of assets, posted just 0.6% gains for the year ending June 2016 compared to the 7.5% return required to meet its payment obligations. This has also brought down its average investment returns over the past two decades to 7%, below its target. About 60% of US pension funds assume a return of 6-7.5%.

The Hoover Institution, a Stanford University think tank, estimates the under-funded US public pension liabilities to be $3.4 trillion. Wilshire Consulting, an investment advisory, estimates that the funding gap of public pension plans in US rose from 23% in 2014 to 27% by mid-2015. As yields stay very low and dependency ratio rises, these liabilities are likely to balloon. Mercer, the consulting firm, estimates that pension deficits of UK’s FTSE 350 companies rose by £21bn to £119bn in just June 2016, almost entirely due to the fall in gilt yields. 

In fact, estimates for the US S&P 1500 companies by Mercer reveals similar disturbing trends. As on August 31, 2016, the estimated aggregate deficit of $570 bn represented a worsening by $166 bn from the $404 bn deficit at the end of 2015. Aggregate funding level of pension plans supported by these companies stood at 77 per cent.

Or as Larry Fink, the Chairman of world’s largest asset manager BlackRock, in his annual letter to shareholders last year, said,

For example, a 35-year-old looking to generate $48,000 per year in retirement income beginning at age 65 would need to invest $178,000 today in a 5% interest rate environment. In a 2% interest rate environment, however, that individual would need to invest $563,000 (or 3.2 times as much) to achieve the same outcome in retirement.

John Authers has said that for “US public employee pension plans, the three percentage points could more than double their deficit, from $1 trillion to $2.5 trillion”. The corollary of this is that these people will now have to cut back on their consumption to save atleast enough to maintain the same level of post-retirement income.

John Mauldin has initiated a series of posts on the looming pensions crisis in the US. As he writes, most of the public pension plans are not fully funded. In fact, as the graphic below shows, the extraordinary monetary accommodation has had a devastating effect, quintupling the the total unfunded liabilities in state and local pensions over the decade. 

Mauldin has dire predictions about the pension plans of city governments in the US, with very bitter consequences for those societies.  

Pension plans cover over 15% of many city budgets, with some having more pensioners than workers. The pension costs of Los Angeles Police Department rose from just 5% of general fund budget in 2002 to 20% today, forcing the LAPD ranks to fall to below 10,000 in 2013 against the required 12500. Similarly, New York today spends more on pensions than it does on building and repairing schools, bridges, parks, and subways. 

The results are hiring freezes which lowers the quality of service delivery and cut backs on the coverage of a variety of services. All this in turn results in higher crime, poor student learning outcomes, declining quality of utility services and so on.

In a different context, India faces an altogether different problem with its pensions. A recent RBI report on household financial savings describes a "ticking pension time bomb". Unlike the US, here it is just that only 7.4% of the working age population is covered by a pension program compared to 65% for Germany and 31% for Brazil. Or unlike UK where private pension investments make up more than 95% of GDP, it is less than 1% in India. The report found that more than 50% of households plan to depend on children for support during old age and 77% do not expect to retire or have not actively planned for retirement.

For India, with its favourable demographic pyramid with rising youth population, this is the right time to encourage savings so as to beneficially smooth inter-temporal financial needs. A recent report by the World Economic Forum reports a global unfunded pension gap of $400 trillion by 2050 for eight countries including India. It estimates India's annual savings gap to grow by 10% and reach $85 trillion by 2050. 

Accounting industry fact of the day

The erosion of values and institutional relationships that underpin capitalism has been a constant theme of this blog. The cozy and entrenched relationship between big businesses and and their auditors is another example. 

Sample this,

All but three of the US’s 500 largest companies use the Big Four accounting firms — Deloitte, PwC, KPMG and EY — and many of them have used the same one for decades. According to data compiled by MSCI, 42 of the big US companies that disclose auditor tenure have hired the same firm for more than 10 years, with an average of 24 years on the job. The UK’s 300 biggest companies are similar — 44 per cent of auditors have been on the job for more than 10 years, with an average tenure of 17 years. Most companies like dealing with the same auditor year after year — fewer things have to be explained, and audits can be completed faster. But critics say these long-lasting relationships can become too cosy and lead auditors to lose their scepticism. They point to UK telecoms group BT, which endured two accounting scandals in nine years while using PwC, whose work is now being probed by the UK accounting watchdog (PwC says it is co-operating, and BT has now changed auditors after 33 years). US lawmakers, meanwhile, have asked for an investigation of KPMG’s work for Wells Fargo during that bank’s fake accounts scandal. KPMG, which has had the account since 1931, has defended its work.

In recent months, the US, UK, and EU have taken some tentative steps to usher in greater transparency into auditing. They include bidding out audit contracts after ten years, changing auditors after 20 years, disclosing auditor tenures, and outlining "critical audit matters" and their resolution. 

Maybe there are compelling reasons, but I struggle to understand why auditors should not be procured by bidding every three years. It is more likely a case of captured regulators choosing the convenience of corporate interests over the concerns of conflicts of interests.

Update 1 (24.09.2021)

Auditing industry fact of the day from India

Bulk of the large listed companies in India prefer the network firms of the big four professional services firms—Deloitte, KPMG, PwC and EY—for statutory audit from among the more than 2,300 statutory auditors in the country, official data showed, indicating their apparent concentration in the premium segment of the audit market. On the other hand, close to 70% of all the statutory auditors in the country contend with just one audit client, as per data available from audit regulator National Financial Reporting Authority (NFRA). The network firms of the big four audited 522 companies in FY19 representing 75% of the market capitalisation of the 5,023 listed companies for which data is readily available, NFRA data showed. This represents about 10% of the listed companies by number. On the other hand, 1,578 auditors audit only one company each accounting for a small fraction of the listed companies.

And this on the. rules, 

Under the Companies Act, listed companies, public companies with paid up capital of ₹10 crore and above, private limited companies with paid up capital ₹20 crore and above and all companies with ₹50 crore and above borrowings from public institutions irrespective of their paid-up capital have to mandatorily rotate individual auditors after five years and audit firms after ten years. The cooling off period between two assignments is five years.

China high-speed rail graphic of the day

FT reports that China is engaged with 18 high-speed railway projects worth $143 bn across the world. A snapshot is below.

I think that unlike the other sectors like telecommunications, roads, power, and even normal railways, this one is unlikely to take off. It is has less to do with Chinese failings or limitations, but the simple fact that in most of these countries high-speed rail, besides being simply unaffordable is also unlikely to even close to being commercially viable.

Interestingly, none of the Chinese contractors, including the large China Railway Corporation (which alone has debt of $558 bn), were among the eleven companies who won the first round of bids for seven contracts worth £6.6 bn tendered out in the first phase of UK's London to Manchester High Speed Rail (HS2) project. The winning bids were from Costain, Strabag, Skanska, Vinci, Balfour Beatty, Carillion, Eiffage, Kier etc. 

The best and worst of Trump matter little to the economy!

I have two long-held views on public policy making. One, even assuming the inevitable cognitive biases like hyperbolic discounting, we live in a world where the immediate takes disproportionate precedence over the long-term. This obsession with the immediate and present may be amplified by modern media and social networks. Whatever the reasons, we end up with ahistorical glorification or demonisation of people, incidents, ideas, and trends that are concurrent with us. Irrational exuberance or pessimism are the result. Such excessive attribution of credit and apportionment of blame blinds us to long-term underlying trends and factors that contribute to such shocks.   

Two, we attribute excessive importance to particular public policy levers in being able to influence long-term economic growth prospects. Governments, be it the Treasury or institutions like the Central Banks, are perceived to hold the keys to the economy. This sentiment becomes egregious during downturns, when governments are blamed for not only contributing to the downturn but also for not doing enough to restore growth. This leaves governments with a need to be seen to be doing something, even when important decision-makers realise the futility of such attempts. 

The best example of both these trends come from US politics and economy in the aftermath of the election of Donald Trump as President. Trump has occupied the centerstage in the cognitive universe of the liberals, demonised as being responsible for the demise of American growth and influence. Important historical trends like rapid convergence of developing countries like China, globalisation, technological disruptions, changing demographics, excessive financialisation, concentration of business activity, corporate rent-seeking, widening inequality, capture of democratic institutions, and so on are swept under the carpet. Trump becomes the punching bag for everything bad about America.  

Much the same applies to both the optimism and pessimism with Trump's economic policies. Matthew Klein points to the graphic below from a presentation by Jason Furman at the recent Peterson Institute Conference on Rethinking Macroeconomic Policy which should explain the need for more balanced assessments.  

In simple terms, the best effects of Trump's "positive" policies (tax cuts, regulatory easing, and public investments) and the worst effects of "negative" policies (protectionism, immigration restrictions, and deficit rise) are equally tiny as to be insignificant over the medium-term. 

The more things change, more they remain the same - macroeconomic policy edition

The breakdown of the Philips curve relationship between inflation and unemployment has been among the major casualties of recent macroeconomic tumult. It has led to demands that central banks replace the current New Keynesian (NK) models involving a demand equation, a policy rule and a Phillips curve to calculate interest rates.

Over the past few years, Roger Farmer and co-authors have shown that an alternative, the Farmer Monetary (FM) Model, which replaces the Phillips curve with a new equation, the belief function, outperforms the NK model by a large margin when used to forecast the US data for the period from 1954-2007. He writes,

The belief function captures the idea that psychology, aka animal spirits, drive aggregate demand. It is a fundamental equation with the same methodological status as preferences and technology. To operationalise the belief function, we assumed that people make forecasts of future nominal income growth based on observations of current nominal income growth... Conventional dynamical systems have a stable steady state that acts as an attractor. The economy will converge to that steady state, no matter where it starts. The FM model does not share that property. Although the economy follows a unique path from any initial condition, the FM model has a continuum of possible steady states and which one the economy ends up at depends on initial conditions. The FM model explains the data better than the NK model because the unemployment rate in US data does not return to any single point... 

The unemployment rate, the inflation rate and the interest rate are so persistent in US data that they are better explained as co-integrated random walks than as mean-reverting processes. The FM model captures that fact. The NK model does not. What does it mean for two series to be co-integrated? I have explained that idea elsewhere by offering the metaphor of two drunks walking down the street, tied together with a rope. The drunks can end up anywhere, but they will never end up too far apart. The same is true of the inflation rate, the unemployment rate and the interest rate in the US data... the NK model is wrong and there has been no stable Phillips curve in the data of any country I am aware ever since Phillips wrote his eponymous article in 1958.

If the recent Peterson Institute conference on Rethinking Macroeconomic Policy is any indication, Farmer and Co may have more work to do. As Matthew Klein writes in FT, the conference co-Chair, Olivier Blanchard, reaffirmed his faith in the Phillips Curve,

I have absolutely no doubt that if you keep interest rates very low for long enough the unemployment rate will go to 3.5, then 3, then 2.5, and I promise you at some point that you will have the rate of inflation that you want.

Translation - keep the monetary accommodation on and unemployment rates will keep going down till they stoke off inflationary pressures "at some point"! Coming as it does from someone who was till recently the Chief Economist of IMF, this is a staggering level of obduracy. And he was leading a conference on "Rethinking Macroeconomic Policy"!

Klein decomposes the Blanchard view in terms of four propositions - jobless rate influences wage bargaining; workers bargain over nominal and not real incomes; workers spend extra money on consumer products, driving up prices; and businesses in the aggregate can raise pay or hike prices. The last three, especially the last, stand on tenuous foundations and limited empirical evidence.

Thankfully, the rest of the conference appears to have been more receptive to change. The most useful reminder for macroeconomists came, as usual, from Dani Rodrik,

The ratio of redistribution to efficiency gains is not only very large, it rises to ridiculous heights as the tax/policy distortion that is removed gets smaller.

He shows that redistribution per dollar of aggregate gain (in favour of the well-off or entrenched) increases sharply as the tax and tariff rates get ever smaller.

Jason Furman called for an end to the debate on whether inequality is good or bad for growth and on its impact on the average of incomes, both irrelevant for policy makers. Instead, he proposes more attention on specific policies that increase or decrease inequality and their impact on indicators that reflect more broader social welfare functions than simple averages (like the impact on specific population categories). His suggestion for developed and developing countries,

In advanced economies a lexicographic framework that focuses exclusively on distributional analysis and then only to growth when the distribution of different policies is the same is generally likely to be appropriate under a broad range of social welfare functions. This is because the distributional effects of many policies are orders of magnitudes larger than the growth effects. In developing economies, however, the scope for policy- and institutionally-induced variations in growth rates is much larger and thus the lexicographic approach is unlikely to be as widely appropriate.

Gita Gopinath puts to rest the orthodoxy about the superiority of freely floating exchange rates and argues in favour of managed floats for emerging market (EM) economies. Channeling BIS economists and Helene Rey, who claim the impossibility of monetary policy independence irrespective of the exchange rate regime, once capital flows are allowed, Gopinath points to the recent work of Obstfeld and Co which shows the persistence of an attenuated version of the trilemma. They show that for EM countries the correlation of a bunch of macroeconomic indicators and global investor risk aversion is lowest for those with intermediate exchange rate regimes.

She also points to the work of Falk Brauning and Victoria Ivashina about outsized influence of US monetary policy cycles on EM economies, driven by global banks, as shown by their finding that over a US monetary easing cycle they experience a 32 per cent loan volume increase.

Weekend reading links

1. Fascinating work by Johnny Miller to highlight segregation in cities using birds-eye view photographs taken using drones.

Such photographs have a powerful way of shaking people out of their comfort zones and generating conversations on the issue of widening inequality and its manifestations.

2. On segregation, nice Citylab interview of Macarthur Prize winner Nikole Hannah-Jones about how racial preferences entrench de facto school segregation in the US. This is spot on,

What we see come immediately out of [Brown v. Board], when you can no longer explicitly use race to segregate schools, is a very adaptive strategy that white Americans have. Suddenly, you take up the banner of race-neutral language that you know will produce the same result. So it becomes about “local control”—saying “our tax dollars shouldn’t go to educate other children,” or “we want a small local school system that only serves our community.” Of course, that community is all white.

In a place like New York City, where you have a great deal of segregation, you have a neighborhood school system for elementary school, which means your kid will go to a neighborhood school. And since neighborhoods are highly segregated, that means your kid will likely go to a school that is also segregated. But then once you get to middle and high school, it is a “choice system” where white kids go to screened schools— schools that have these apparently race neutral screens, but where you have to have a portfolio, or where you have to take a test to get in. What remains the same is that white parents are going to get access to the best education in a public system. They’re going to get access to disproportionately white schools, and they will wield an array of tools to do that. So if the neighborhood that those white parents live in is white, they want neighborhood schools. If the neighborhood school that those parents are near is black, then they want choice. So people will say they don’t want bussing, if their neighborhood school is white. If the neighborhood school is not white, they’ll bus their kids an hour away to get to a white school.

At some level, the difference between Trump and the liberals are not as glaring as you think. Trump acts and does so with brazenness what most others secretly nurture inside or share across the dinner table.

3. I have blogged earlier about this, but worth repeating this time with Andy Mukherjee's graphic on the critical role of excise duty of petroleum products in India's fiscal consolidation.

The flip-side is that as global economy steams on, the only direction for oil prices may be upwards. This would set in motion the exact reverse dynamics - slowly roll-back the duties to absorb the price increases. With several elections round the corner in the lead-up to 2019, the political compulsions will be overwhelming.

One more reason to hold pause on any thoughts of a fiscal expansion.

4. Livemint editorial captures the wrinkles in India's mobile phone manufacturing surge,

India has also shown impressive growth in manufacturing smartphones, almost tripling the value of output from Rs18,900 crore in 2015-16 to Rs54,000 crore in 2016-17. This is expected to increase to Rs94,000 crore in 2016-17. But all these “Made In India” smartphones, including those from home-grown companies like Micromax and Karbonn, are only assembled in India. Smartphone companies import semi-knocked down (SKD) units, in which all the key high-value components are already soldered. The value addition happening in India was only 6.1% of the smartphone’s value in 2016... In the 2015-16 Union budget, the government increased the differential excise duty structure for mobile phones from the earlier 5% to 11%, which gave domestic manufacturers a benefit over imported phones. This move has managed to flip the share of imported mobile phones from 69% in 2015 to 33% in 2016.

5. FT points to an interesting October Curse with financial markets, captured by the graphic below.

6. I have never been a great admirer of corporate India's dynamism and claims that it can propel India to become an economic super-power. And it is not just because of the questionable corporate governance standards, but also because of its relative failure to produce truly world-class innovations despite numerous opportunities.

The telecoms sector always gets held up as the poster child of what unshackling India's corporate sector from the License Raj can yield. While the explosive growth of cellular telecoms over the past decade and half have been truly transformative, there is also a less gratifying side.

Here is Sundeep Khanna's lament in Livemint about the sorely deficient R&D spending of India's telecom companies,

It is safe to say that just as information technology (IT) services were the success story of the 1990s, telecom services have been the defining story of Indian business in the first decade of the 21st century. Sadly, for all its growth and penetration, the sector has thrown up little by way of research and development. Not a single breakthrough technology in telecom has come out of India, which should have served as the perfect crucible for experimentation... In the list of 50 top applicants that filed under the World Intellectual Property Organisation’s Patent Cooperation Treaty (PCT), a list dominated by telecom companies like ZTE Corp., Huawei Technologies and Qualcomm Inc., there isn’t a single Indian name. Even worse, a report by Thomson Reuters last year named Samsung, Huawei, LG, State Grid Corp. of China, ZTE, Qualcomm, Ericsson, Sony, NTT and Fujitsu as the top 10 global innovators in the telecommunications industry for 2015. No mention again of an Indian telco.

All the ingredients that experts say are needed to fuel research and development in the sector have been present in India, a large and rapidly growing market, demanding customers, profits aplenty, a phalanx of financing options and intense competition that should logically have spurred innovation. What’s more, it was evident to anyone who cared to look that in a business driven so much by technology, the barriers to entry could collapse overnight as it happened when Reliance Jio made its moves a year ago. The bloodbath that has ensued hardly comes as a surprise.

7. The use of satellite data on night lights in economic data analysis is on the rise. Two good articles - one on the use and limitations of the use of such data, and another on why the NASA data on India may not exactly represent what is being claimed.

8. FT documents the property market rebound in China,

Urged on by Beijing, 38 per cent of all bank loans issued in the 12 months to August were home mortgages, according to official data, and local governments purchased 18 per cent of all residential floor space sold last year as part of a drive to provide affordable housing, according to estimates by E-House China Research Institute. The result has been another heady boom in construction. Rome was not built in a day, but based on residential floor area completed last year, China built the equivalent of a new Rome about every six weeks... For China’s domestic economy, the world’s largest at purchasing-power parity, property investment directly contributed 10 per cent to GDP in 2016. When manufacturing sectors like steel, cement and glass and retail sectors like furniture and home appliances are included, the share is at least 20 per cent...

A survey by FT Confidential Research, an independent research service owned by the Financial Times, found that 32 per cent of families own at least one home that is vacant. An estimated 50m homes, or 22 per cent of the total urban housing stock, were vacant in 2013, according to the most recent data from the China Household Finance Survey led by Li Gan, economics professor at Texas A&M University.

9. Madhya Pradesh is experimenting with a new approach to Minimum Support Price (MSP). Instead, under the Bhavantar Bhugtan Yojana, of government itself making the purchases at the MSP, the farmers will be reimbursed the differential between the MSP and their sale price. The experiment is being tried out for a period of two months for eight crops. The sales will have to be done in registered agriculture markets and subsidy will be transferred to the farmer's bank account. To prevent traders from artificially suppressing the market, the government will determine an average sales prices, not only based on the prices in the State but also in two neighbouring states.

This is a likely more efficient approach. But like with all such reforms, given the several pathways to game the new system, the success lies in effective implementation.

10. Finally, the latest adventure in kritarchy comes by way of the Supreme Court's decision to ban the sale of fireworks in Delhi during the Diwali festival.

This is clearly grandstanding judicial activism since there is at least as much or more evidence that the important contributors to air pollution in Delhi lie elsewhere - burning of crop waste in Punjab and Haryana, construction activity and vehicle emissions.

India R&D spending fact of the day

From Janak Nabar in Livemint,

Huawei’s R&D expenditure (around $6.5 billion) is about the same or more than that of Indian industry, while Microsoft spends (around $12 billion) about the same as the Indian government.