Veritas propter investigationem [Truth through research]
TitleMatching Contributions for Pensions – A review of international experience (2013)
AuthorsRichard Hinz
 Robert Holzmann
 David Tuesta
 Noriyuki Takayama
InstitutionWorld Bank
TopicsTier 2 pensions
 Defined Contribution schemes
 Participation rates
 Pension reform evaluation
 Retirement income adequacy
Date Published2013
Date posted on PR16 Dec 2014
Hinz, R Holzmann, R Tuesta, D Takayama, N, (2013). Matching Contributions for Pensions – A review of international experience (2013) World Bank,

PensionReforms’ summary and comments

This 356 page, 2013 publication is a collection of essays on aspects of nine countries’ Defined Contribution arrangements.  This abstract will deal with the report’s own summary and conclusions.  PensionReforms will deal separately with some of the publication’s chapters on specific countries’ experiences.


“After decades of reform and innovation ranging from individual mandates to creative forms of tax incentives, coverage of pension systems often remains at less than half of the economically active population - in part because the incentives for contributing to pension savings are often obscure or irrelevant to those who need this protection the most: young and low-income individuals with irregular earnings and little attachment to the formal labor force.  Traditional forms of tax incentives have little relevance to those not paying income taxes, and participation mandates are difficult to enforce in the informal sector.”


The report looks at the experience of the countries with that match employees’ contributions to improve participation levels and outcomes.  The authors suggest that “matching defined contributions schemes are gaining popularity in both rich and poor countries…[but that] it remains far too early to develop firm conclusions or policy guidance…”.  The experience thus far “suggests that matching is moderately effective in increasing program participation but not generally measurably effective in raising contributions and thus benefit levels.”


The report suggests that “…other interventions – which are increasingly guided by lessons from behavioral economics and finance – may prove to be more effective and typically cost much less…”  Much more work is needed.


The report emphasises the need for “broad pension coverage and adequate levels of income protection in old age” but notes that “…in the vast majority of countries, less than half the working population is covered by a formal pension scheme.”  Despite compulsion, “many middle-income countries have seen coverage rates decline in recent decades…”


This report suggests that the World Bank’s review of the role of social pensions (Holzman, Robalino and Takayama 2009 (reviewed here) dealt with the state’s responsibility at Tier 1.  PensionReforms thought that the 2009 report was “a step in the right direction”.  However, because the 2009 report dismissed universal pensions at Tier 1 as “likely to be sub-optimal” without looking in detail at any country’s experience with a Universal Tier 1 pension, this latest review necessarily rests on an inadequate foundation.


The 2009 report concluded that compulsory Tier 2 schemes seemed not to be ‘working’ in all but a handful of developed countries.  It described the “limited coverage of mandatory pension systems” that seemed to PensionReforms to summarise what was wrong with the World Bank’s original 1994 prescription in Averting the Crisis. (see here).  A compulsory scheme with limited coverage is one definition of ‘not working’.


From that starting point, this latest report looks to see what can be done about Tier 2 and above.  Solving the problem at Tiers 2 and 3 may need “stronger incentives for participation in formal pension and saving schemes”.  The report reviews the on-going use of “matching defined contribution” (MDC) schemes but is not intended to “formulate or articulate a World Bank policy position”.  Partly, that’s because the “evidence is far too limited to support such an effort”.


Extending pension coverage isn’t the only objective.  There are two further aims - “…reducing informality, and increasing fiscal efficiency.” The eventual objectives are, at a high level, to “correct market failures and to redistribute income….a third has been added in recent years: to correct the behavioural limitations of individuals.”


The report suggests a possible fourth reason: “…to correct government failure itself – by, for example, redesigning social insurance programs that do not achieve sufficient coverage due to poor design or implementation.”


PensionReforms thinks that the report’s overall findings are unsurprising and emphasise the gaps in the World Bank’s earlier 2009 report cited above.  The only way a government can ensure that age pensions are both adequate and achieve acceptable coverage is to offer a Universal Pension at Tier 1.  This avoids the need to worry about ‘informality’ in the workforce on pension grounds alone (there are other reasons to worry about that though); fiscal efficiency is assured; regulatory walls around tax-favoured and compulsory Tier 2 schemes can be dismantled and the state can avoid intervening in the remuneration decisions of employers (with an employer match contributions).  The state should be worrying about the things it can change rather than trying to second-guess citizens who choose not to do what the state has decided they should do (in their own best interests, of course).


If a Universal Pension is dismissed as “likely to be sub-optimal” then public policy is inevitably forced into a discussion about how to deal with those who fall through the cracks, actively avoid participation or don’t know what’s good for them.  Around the world, and especially in low and middle-income countries, the numbers in those groups seem to be growing.  MDC schemes can’t fix that problem, just as compulsory Tier 2 schemes can’t refinance the state’s obligation at Tier 1.  (File size 1.75 MB; 356 pp) 710