Discussion papers and reports

Discussion papers analyse the theoretical and empirical aspects of a particular topic in detail. Special reports typically cover a wider subject area.

This page lists publications and provides an abstract or an executive summary together with a link to the document.

Using U.K. microeconomic data, the authors analyze the empirical determinants of voluntary annuity market demand. The authors find that annuity market participation increases with financial wealth, life expectancy and education and decreases with other pension income and a possible bequest motive for surviving spouses. The paper shows that these empirically-motivated determinants of annuity market participation have the same, quantitatively important, effects in a life-cycle model of annuity demand, saving and portfolio choice. Moreover, reasonable preference parameters predict annuity demand levels comparable to the data, thereby questioning the conventional wisdom that limited annuity market participation is a puzzle to be explained.

In many countries, pension funds based on individual accounts have been affected by high operating costs. Contract theory helps to unravel the nature of such problems: managers of pension funds have strong incentives to manipulate market expectations about their capacity through wasteful activities (e.g. promotion). Thus, competition among pension funds entails efficiency loses, due to pension savings attraction efforts, as well as gains, related to investments in asset management skills. Regulations capping fees or costs of pension funds worsen market inefficiency, while a public pension fund competing with private ones improves (at least weekly) it. Taking into account political and commitment constraints affecting public institutions, a quasi-competitive pension scheme - centralizing contribution collection, auctioning the right to manage raised money to competitive fund managers, and affording an opting out choice to households -Pareto-dominates (at least weekly) the market of pension funds.

The theory of equalizing differences suggests that employer provided pension benefits should be compensated by reduced wage benefits for an employee's given productivity potential. This paper presents an empirical analysis of compensating wage differentials for occupational pension scheme benefits in the UK using the newly available English Longitudinal Study of Ageing. The data allows us to differentiate between Defined Benefit (DB) and Defined Contribution (DC) schemes and to consider different measures of pension benefits based on current contributions and changes in accrued pension benefits rights. In our preferred specifications we find evidence for perfect compensating wage differentials for both occupational DB and DC pension scheme benefits.

This paper sets out the economic analytics of pensions. After introductory discussion, successive sections consider the effects of different pension arrangements on labour markets, on national savings and growth, and on the distribution of burdens and benefits. These are controversial and politically highly salient. While we are open about expressing our own views, the main purpose of the paper is to set out the analytical process by which we reach them, to enable readers to form their own conclusions.

UBS PRP Discussion Paper No.40(FMG 562)
Barr, Nicholas "Pensions: Overview over Issues"
May 2006

Many countries face increasing fiscal problems financing pensions in the face of population aging. There is controversy about the underlying economic theory, about the extent of the problem, and about the best mix of policies to protect old-age security. This paper establishes the areas of debate: gives thumbnail descriptions of pension arrangements in different countries; discusses the main analytical and empirical issues relevant to thinking about pension design; and assess a range of policy directions. The main conclusions are that what matters most is effective government and economic growth; that the debate between pay-as-you-go and funding is secondary; that good pension schemes can take many forms; and that there is a problem in financing pensions, but not a crisis.

We investigate whether a rare event (like the default of the annuity provider) can explain the annuity market participation puzzle. High risk aversion is needed to change behavior in the presence of such a disastrous shock but higher risk aversion also makes annuities more valuable. Therefore, these rare events are unlikely candidates to explain the low take-up of voluntary annuities.

This essay describes the current debate on reforming Social Security in the US, along with a brief description of how the program works. Along the way it comments on the quality of some reform proposals as well as their political standing. Where issues are similar, some inferences are drawn for the UK.

This paper studies optimal intergenerational transfer policy under stochastic labor income and capital returns. It has implications for Social Security, government tax and debt policy, and DB pension funds. A stylized two-period overlapping-generations model is developed where a central planner implements pay-as-you-go transfers. I allow for autocorrelation in the labor income and skewness in the capital return and calibrate the model parameters to US data. I show that state-contingent transfers facilitate intergenerational risk sharing in a way that is similar to portfolio insurance using put options. That is, the working generation provides downside risk insurance to the old on their savings. In addition, when no riskfree asset is available, these transfers improve utility by substituting for this missing asset. I further find that imposing an incentive constraint for the working generation has little impact when transfers also have this substitution role, but it causes the transfer scheme to collapse to the zero-transfer scheme when a risk free asset is available.

We investigate the impact of owner-occupied housing on financial portfolio and mortgage choice under stochastic inflation and real interest rates. To this end we develop a dynamic framework in which investors can invest in stocks and bonds with different maturities. We use a continuous-time model with CRRA preferences and calibrate the model parameters using data on inflation rates and equity, bond, and house prices. For the case of no short-sale constraints, we derive an implicit solution and identify the main channels through which the housing to total wealth ratio and the horizon affect financial portfolio choice. This solution is used to interpret numerical results that we provide when the investor has short-sale constraints. We also use our framework to investigate optimal mortgage size and type. A moderately risk-averse investor prefers an adjustable-rate mortgage (ARM), while a more risk-averse investor prefers a fixedrate mortgage (FRM). A combination of an ARM and an FRM further improves welfare. Choosing a suboptimal mortgage leads to utility losses up to 6%.

We solve a model with incomplete markets and heterogeneous agents that generates a large equity premium, while simultaneously matching stock market participation and individual asset holdings. The high risk premium is driven by incomplete risk sharing among stockholders, which results from the combination of borrowing constraints and (realistically) calibrated life-cycle earnings profiles, subject to both aggregate and idiosyncratic shocks. We show that it is challenging to simultaneously match aggregate quantities (asset prices) and individual quantities (asset allocations). Furthermore, limited participation has a negligible impact on the risk premium, contrary to the results of models where it is imposed exogenously.

This paper uses data from the British Household Panel Survey to shed further light on the fall in spending at retirement (the “retirement-consumption puzzle”). Comparing food spending for men retiring involuntarily early (through ill health or redundancy) with spending for those who retire voluntarily, it finds a significant fall in spending only for those who retire involuntarily. This is consistent with the observed fall in spending being linked to a negative wealth shock for some retirees. Evidence on psychological and financial well-being also indicates that the retirement experience of involuntary retirees is very different to that of voluntary retirees.

The paper presents an analysis of the impact of pension plan funding on workers’ saving and portfolio behaviour. It shows that the impact of pension plan funding and asset allocation on the economy’s technology choices depends upon the constraints facing worker’s in the capital market. The failure of equivalence propositions between defined benefit and defined contribution pension plans derives from the existence of borrowing and short-sales constraints. We investigate how firms’ capital structure choices interact with pension plan funding, both when corporate debt is riskless and risky. We make predictions about how, in the presence of leverage, the benefit level and financing of the pension plan affects technology choices and aggregate risk premia. The impact of leverage on these variables is shown to be reversed if if non fully-funded pensions themselves are put at sufficient risk.

As a result of population ageing and declining fertility, the UK state pension system is unlikely to remain viable in the very long run without a steady inflow of young immigrant workers from abroad. However, with prudent economic management and continuing economic growth, immigration requirements can be contained and modest real increases in pensions are a possibility. Beyond 2020, further ageing of the population will lead to fiscal pressures and the need for remedial measures such as the raising of the state pension age. Higher economic activity rates among older people, including deferred retirement, will to some extent ameliorate but not eliminate these pressures. If fertility picks up over the next few years, this will also help, but not until after 2030. Without favourable economic growth, the fiscal problems will appear much sooner and could lead to cuts in pensions or to significantly higher contribution rates.

Large sections of the UK population are failing to make adequate provision for their retirement, and one area where provision is notably poor is amongst people working for small and medium sized businesses. We use information gathered from interviews with individuals from a wide range of organisations active in the pensions market for these companies to shed light on the particular barriers to pension scheme participation in this sector in which over 40% of the working population is employed. We find that many finance directors are sceptical of the benefits of providing pensions for their employees and deliberately structure their pension schemes to avoid high participation rates. We also find that financial advisers and pension providers are reluctant to promote pensions in companies where they perceive the management to be unsupportive and where there is no clear profit margin. We suggest one way in which management could be motivated to encourage pension scheme participation amongst their employees.

We calibrate a life-cycle model with uninsurable labor income risk and borrowing constraints to match portfolio allocation and wealth accumulation profiles of direct and indirect stockholders in both taxable and tax-deferred accounts. Tax-deferred accounts generate an increase in wealth accumulation that is larger for wealthier households. Furthermore, while the cost of following a fixed contribution rate over the life cycle is small, the optimal rate can differ substantially across households, and the welfare losses from choosing the wrong one can be substantial. Finally, the welfare gain from having access to a tax-deferred account ranges from less than 0.1% to 11.5%, depending on the preference parameters.

Current approaches to asset-liability management employ a sequence of distinct procedures to value liabilities and determine the asset allocation. First, a discount rate that is usually dictated by accounting standards is used to value liabilities. Second, the asset allocation is determined by maximizing some objective function in the surplus of assets over liabilities, taken as given the valuation of liabilities. We introduce a model that allows for the joint valuation of liabilities and the determination of the optimal asset allocation using discount rates that appropriately reflect default risk. We focus on the case of a defined benefit pension plan.

UBS PRP Discussion Paper No.26(FMG 505)
Cocco, Joao and Paula Lopes "Defined Benefit or Defined Contribution? An Empirical Study of Pension Choices"
July 2004

We empirically study individual pension choice between two different defined benefit (DB) plans and a defined contribution (DC) plan. The DB plans differ in their contribution rates and in the way retirement benefits are calculated, as a proportion of final salary or as a proportion of lifetime earnings. We relate labor income characteristics to the choice of pension plan. Among other determinants of pension choice, we find that: (i) individuals who face higher income growth are more likely to choose DB final salary plans, and less likely to choose the DC plan; (ii) individuals who face higher earnings volatility are less likely to choose DB final salary plans; (iii) individuals with higher earnings are more likely to choose either the DC or the DB final salary plan. These results constitute evidence of self selection of individuals into different pension plans, an important issue for pension fund providers and for those involved in pension reform.

In a model of career concerns for experts, when is a principal hurt from observing more information about her agent? This paper introduces a distinction between information on the consequence of the agent's action and information directly on the agent's action. When the latter kind of information is available, the agent faces an incentive to disregard useful private signals and act according to how an able agent is expected to act a priori. This conformist behavior hurts the principal in two ways: the decision made by the agent is less likely to be the right one (discipline) and ex post it is more difficult to evaluate the agent's ability (sorting). The paper identifies a necessary and sufficient condition on the agent signal structure under which the principal benefits from committing not to observe the agent's action. The paper also shows the existence of strategic complementarities between information on action and information on consequence. The results on the distinction between action and consequence are then used to interpret existing disclosure policies in delegated portfolio management. In particular, they are consistent with hitherto puzzling evidence that mutual funds systematically overperform pension funds.

We show that a life-cycle asset allocation model with liquidity constraints and realistically calibrated uninsurable labor income risk rationalizes the asset allocation puzzle of Canner, Mankiw and Weil (1997). Based on empirical estimates of the correlation between stock returns and individual earnings, labor income is a closer substitute to long-term bonds than to stocks. As a result, more risk averse investors hold a smaller proportion of their risky portfolio in equities. Moreover, this explanation is consistent with the recommendation that younger households should be more heavily invested in stocks than older households.

This paper presents a model of the interaction of a company's financial and real investment decisions with the financing of its defined benefit pension plan. The pension plan deficit is a debt of the company, with explicit funding requirements and priority in the event of company insolvency. Pension plan deficits and options on future deficits and surpluses affect investment incentives as does the size and composition of company debt. We illustrate the incentives for the firm to pay dividends rather than fund the pension plan and the general incentives to overfund the pension plan. We also illustrate the impact of pension benefit insurance and minimum funding requirements.

UBS PRP Discussion Paper No.22 (FMG 486)
Gregory, Alan and Ian Tonks: "Performance of Personal Pension Schemes in the UK"
March 2004

This paper examines the performance of personal pensions (exempt unit trusts) in the UK 1980-2000. Unitised personal pension schemes are a type of mutual fund that is constituted as a contractual savings scheme, whose value can only be accessed at retirement. By studying the performance of these schemes we are able to assess the role of illiquidity in retail savings products. The paper examines those personal pension schemes that invest predominantly in UK equities, and first reports on the growth in personal pension schemes over this twenty-year period. The paper then assesses the performance of these pension funds relative to various asset pricing benchmarks, including a four factor benchmark that allows for momentum in stock returns, and allowing for market timing and conditioning on macroeconomic variables, and finds that average performance is not significantly different from zero. The paper goes on to examine persistence in performance of these pension schemes and identifies negative persistence at short horizons, but at time-intervals of six months to one year finds significant positive persistence, though this positive persistence weakens at longer time intervals.

With a move away from up-front charges following the introduction of stakeholder pensions, consumers are no longer penalised for lapsing on many long-term savings policies. Nevertheless, persistency rates may still provide an (imperfect) indicator of sales quality and provide some information on how consumers are building up savings for the longer-term. Furthermore, persistency is an increasingly important issue for financial providers and the profitability of stakeholder-friendly products. This paper uses aggregate persistency data and survey data from the British Household Panel Survey to address three key questions: What drives persistency rates among different groups in the population? To what extent does non-persistency appear to reflect poor sales and advice, rather than events in consumers’ lives that were not predictable at the time of sale? Are there any messages that could be given to the industry or to consumers to help raise levels of persistency?

The authors demonstrate that a lifecycle model with realistically calibrated uninsurable risk to labour income and moderate risk aversion can match both stock market participation rates and asset allocation decisions conditional on participation. The key ingredients are Epstein-Zin preferences, some heterogeneity in risk aversion and a fixed stock market entry cost. Households with low risk aversion smooth earnings shocks with just a small buffer stock of assets and do notm invest in equities. The marginal stockholders are more risk-averse and as a result they do not invest their portfolios fully in stocks.

This paper solves an empirically parameterised model of optimal household demand for nominal and inflation-indexed annuities. The model incorporates mortality, inflation and real interest rate risk. It draws some surprising predictions.

Nominal annuities improve welfare even when sold at the empirically paramerised cost, above fair value. Real annuities improve welfare even more at the fair price but the welfare gains turn negative at the empirically parameterised level of the annuity premium.

Using the UK wealth distribution for simulations, it turns out that an annuity costs a lot of money compared to the total accumulated assets held by a typical household. In other words, demand is low because many households cannot afford to enter the market.

Among households who do buy them, simulated demand for annuities tends to match the actual demand reported in the Family Resources Survey.

The authors investigate the macroeconomic and welfare effects of DB and DC systems, and the effects of incremental reform within a given system, using an equilibrium life-cycle model with incomplete markets and heterogeneous agents.

Using calibrations to illustrate a tax-financed DB system's trade off between efficiency and redistribution, they show that social welfare is maximised for small positive levels of DB.

On the other hand, steady-state within-DC system comparisons reveal that a zero DC tax rate maximises social welfare.

This paper investigates the estimation of long-term VaR. It also suggests a simple approach to the estimation of long-term VaR that avoids problems associated with the square-root rule for extrapolating VaR, as well as those associated with attempts to extrapolate day-to-day volatility forecasts over longer horizons.

Using the Financial AIDS model, the author examines the allocation of UK personal sector wealth across five asset categories: net financial wealth, housing and durable assets wealth, state pension wealth, private pension wealth, and human capital.

Besides total wealth and returns, variables relating to capital market imperfections and to demographic, labour market and cross-sector spillover effects also turn out to be significant. The adjustment of portfolio weights to shocks is very slow, taking up to 21 years for some asset categories.

As a result of population ageing and declining fertility, the UK state pension system is unlikely to remain viable without a steady inflow of young immigrant workers from abroad. The author shows that up to 500,000 immigrant workers pa. will be needed to save the state pension system.

Other things equal, a net inflow of immigrant workers is required whenever: real pensions growth exceeds 2.5% p.a.; the number of retirees exceeds 660,000 p.a.; the number of deaths falls below 570,000; real wage growth is below 1.5% p.a.

This article is based on written evidence presented to the House of Lords Economic Affairs Committee of Enquiry into ‘Aspects of the Economics of Ageing’.

In examining the financial system prerequisites needed for the successful delivery of funded private pensions, the author looks at the financial instruments and investment strategies required during the accumulation and decumulation phases. The context for this analysis is the UK, i.e. a specific developed economy with a mature pension system. A variety of detailed lessons are learned, and these may be used tom inform the debate in developing countries that are in the process of undertaking pension reform.

The author describes three types of political risk which have affected the UK second-pillar state pension (SERPS) and shows that these have resulted in a drop in the real IRR from 5% to 1.5% over 25 years. The current government replaced SERPS with S2P, designed to benefit low-paid workers the most. That constituency's prospective IRR has risen to 6.2% as a result of S2P, plus Mimimum Income Guarantee (MIG) and Pension Credit. The author demonstrates that even well-off pensioners will become eligible for MIG and questions whether this benefit can continue in its current form.

The UK first-pillar pension's IRR has also declined, by 3%, as a result of the decision to index to prices instead of earnings. Personal pensions have also been adversely affected by political risk, though less. Given these risks to DC schemes and the fact that company final-salary schemes have all but closed to new members, the author concludes that it is hard to see how UK workers will be able to guarantee their retirement income.

The authors use contract theory to analyse such aspects of pension governance as: responsibility for asset management; the role of trustees; asset allocation; and funding. The residual claimant of a DB and DC plan is, respectively, the sponsor and the beneficiaries. The authors show that, with complete contracting, governance is neutral and optimality of DB or DC is determined by relative risk tolerance. The less risk-averse party should be the residual claimant.

Once "vigilance" becomes non-contractible, the party with the comparative advantage may not have an incentive to take on that role. Under these circumstances, a DB plan should: (i) assign more vigilance responsibility to the sponsor, less to the beneficiaries; (ii) rely relatively less on trustees; (iii) choose trustees that are professional experts rather than caring insiders; (iv) assign asset allocation rights to the sponsor rather than the beneficiaries. The opposite is true for DC plans.

The author reviews the current system of pension provision and analyses the reforms since 1980. He examines the framework (legal, regulatory and accounting) for occupational pension schemes and assesses the types of risk and returns from membership of DB and DC schemes. This is followed by an empirical analysis of historical management and investment performance of pension fund assets; and a discussion of the review of institutional investment in the UK conducted by Paul Myners.

Using stochastic modelling, the author demonstrates that the best investment strategy for the accumulation phase of a DC pension plan is one that limits the range of returns that are credited to the plan member's account. In particular, with-profit accumulation programmes, which use a smoothing fund to smooth out returns over time, dominate unit-linked accumulation programmes.

For the decumulation phase, the authors show that it is hard for an investment-linked programme to beat the income and security provided by a standard annuity but, again, with-profit programmes dominate unit-linked programmes.

Return smoothing is therefore argued to be a valuable feature of any long-term investment programme, both during the accumulation and the decumulation phases, with important implications for the design of Sandler "stakeholder" products.

The author analyzes the Myners Report and concludes that it will change asset selection, the role of the fund manager and performance measurement.

Asset classes will be selected so as to match liabilities in terms of correlation and volatility. Every pension scheme will have a scheme-specific funding standard that reflects the maturity structure of liabilities.

A hierarchical relationship will develop between investment advisor, actuary and fund manager, where the investment advisory function will assume primacy over actuarial and fund management functions.

Liability-driven performance measurement and attribution will replace the current UK framework. Passively managed components of the pension fund will be judged on the costs of implementation. Only the performance of the surplus assets will be measured on a conventional basis.

The authors construct a time series of annuity rates in the UK from 1957 to 2002, and:

Using data on annuity and other security returns, the paper simulates replacement ratios and builds up a frequency distribution of the pension replacement ratio of a UK individual.

These frequency distributions illustrate the risk in the pension replacement ratio faced by an individual member of a typical DC pension scheme.

Deterministic lifestyling (the gradual switch from equities to bonds according to preset rules) is a popular asset allocation strategy during the accumulation phase of a DC pension plan and is designed to protect the fund from a fall in the stock market just prior to retirement.

The authors show that this strategy can be highly suboptimal because it fails to take account of the degree of risk aversion and the correlation over time between the plan member's salary and the stock market.

It is dominated by a dynamic asset allocation strategy, which the authors call "stochastic lifestyling", that does take these factors into account -- and even by a static strategy that takes these factors into account.

This paper considers the choices available to a DC pension plan member, at the time of retirement, for conversion of his pension fund into a stream of retirement income.

The authors compare the purchase of a conventional life annuity (bond-based) with distribution programmes with various degrees of equity exposure during retirement. The residual fund at the plan member's death can either be bequested to his estate or exchanged for survival credits while alive.

The authors find that the most important decision, in terms of cost to the plan member, is the level of equity investment; and that the optimal age to annuities depends on the bequest utility and the investment performance of the fund during retirement.

Housing and pension wealth are shown to be important determinants of personal sector consumption and retirement behaviour in the UK. Housing and state pension wealth have a positive effect on consumption, while private pension wealth promotes greater savings. Greater private defined benefit pension wealth encourages earlier retirement, while greater defined contribution pension wealth has the effect of delaying retirement. State pension wealth appears to have no effect on the retirement decision. Other variables relating to income, labour market and demographic status and spillovers from other sectors are also shown to be important. The consumption equation forecasts the late 1980s boom and the early 1990s slump in the UK better than other models that disregard housing and pension wealth. A particularly important cause of the boom was the huge private pension fund surplus that accrued as a result of the stock market boom of the 1980s.

This paper proposes new performance decomposition measures that allow us to analyze the sources of returns on the international equity holdings of a large cross-section of UK pension funds. Our results suggest that the pension funds earned negative returns both from international market timing and from selecting stocks within individual foreign regions. The average fund under performed a passive global equity benchmark by 70 basis points per annum. This is substantially greater than UK pension funds' underperformance in their domestic equity market. We discuss the implications of these findings for theories of informational asymmetries in international stock markets.

Despite being largely unconstrained in their investment decisionsons, we find evidence of clustering in the performance of a large cross-section of UK pension fund managers around the median fund manager. We explain this finding in terms of: the predominance of a single investment style (balanced management), the fee structures and incentives operating in the UK pension fund industry to maximize relative rather than absolute performance, the high concentration in the UK pension fund industry and the low turnover of fund managers. Fund size appears to be the only variable that can account for an important fraction of the cross-sectional variation in measured performance.

This paper analyzes the international equity holdings of a large parcel of UK pension funds. We find considerable evidence of market timing activity, as illustrated by the funds' decision to scale back their investments in the US stock market during the 1990s. To explain this we model portfolio weight dynamics as a function of time-varying conditional moments. We find that a substantial part of the evolution in portfolio weights is explained by time-varying conditional expected returns, volatilities and covariances with domestic equity returns. Consequently, controlling for the effect of state variables that track time-variations in investment opportunities significantly affects estimates of returns from international market timing. Our estimates suggest that the portfolio movements that were orthogonal to such state variables accounted for a net loss of 0.2 percent per annum for the average fund.

This paper examines persistence over time in the performance of fund managers responsible for making the investment decisions of UK pension funds. Previous work on UK pension funds found little evidence of fund manager persistence, but we argue that this may have been due to survivorship bias in the construction of these data samples, which may have disguised true persistence. Using a large sample of pension funds over the period 1983-97 in which there is less survivorship bias, we find strong evidence of persistence in abnormal returns generated by fund managers over one year time horizons.