City Research Online

The Annuity Puzzle: Insights from Recent UK Pension Developments

Findlater, A. (2023). The Annuity Puzzle: Insights from Recent UK Pension Developments. (Unpublished Doctoral thesis, City, University of London)


Annuity sales in the UK fell sharply following changes to pensions regulation referred to as “Freedom and Choice in Pensions” (or more loosely as pension freedom or the pension freedoms). This created an opportunity for novel research into a long-standing academic question: the annuity puzzle (- why consumers needing lifetime income tend not to purchase annuities voluntarily). This thesis uses mixed methods to create a rich picture of the UK pensions market as it responded to regulatory reform, investigating how this affected annuity demand and making recommendations to improve consumer outcomes.

The research covers analysis of annuity pricing, a review of publicly available material providing guidance about retirement income, interviews with 48 pension practitioners, and the investigation of product solutions using a parsimonious financial model. Computer-based Natural Language Processing (NLP) was used to strengthen and extend the research into retirement income material, and that approach was then also applied to transcripts of the practitioner interviewees.

Annuity pricing was based on case-specific mortality assumptions, essential in a market where cases are routinely underwritten using health-related indicators, building on previous studies based on generic mortality. The conclusion was that pricing remains fair. This refutes one possible hypothesis: that the reduction in sales might have been driven by a worsening in price, connected with an observed reduction in the number of providers.

The analysis of retirement income material finds that it conveys the message that annuities are expensive and inflexible, whereas the main alternative product (income drawdown from an invested fund), is presented much more positively and is positioned as fitted to the modern world. The material reinforces factors which the academic literature suggests are negative for annuity purchase, in particular by emphasising investment aspects and downplaying the challenges of generating secure lifetime income. These findings arose from my own thematic analysis of material, and were then validated by NLP modelling.

The interviews reveal sharp disagreements about whether recent reforms were beneficial and how outcomes could be improved, but with general agreement that the market environment was not prepared for the reforms. Lack of consumer engagement and understanding are seen to be critical problems: consumers who are ill-fitted to make good use of freedom and choice have had that thrust upon them, but with inadequate support in using them. In terms of product solutions, there was general agreement that innovations to fit the new environment have been slow to embed themselves in the market. My analysis of simple product combinations using a parsimonious financial model suggests that key consumer objectives of secure lifetime income, benefits on death and flexibility could be delivered with
straightforward combinations of simple products, in a way which was not commonly pointed to in the interviews and indeed which was in some respects directly contrary to opinions expressed.

My main recommendation for improvement tackles this by requiring good default strategies to be made available to all, as an obligation on firms providing defined contribution pension schemes. This preserves consumer freedoms through the availability of opt outs, while ensuring that disengaged consumers who do not exercise active choice end up in good retirement income strategies.

Publication Type: Thesis (Doctoral)
Subjects: H Social Sciences > HG Finance
Departments: Bayes Business School > Bayes Business School Doctoral Theses
Bayes Business School > Finance
Doctoral Theses
[thumbnail of Findlater thesis 2023 PDF-A.pdf] Text - Accepted Version
This document is not freely accessible until 31 January 2027 due to copyright restrictions.


Add to AnyAdd to TwitterAdd to FacebookAdd to LinkedinAdd to PinterestAdd to Email


Downloads per month over past year

View more statistics

Actions (login required)

Admin Login Admin Login