Personal learning accounts (PLAs) are seen as an important tool to support efforts by the government to increase the skill levels of the UK population by providing financial incentives for individuals and employers to invest in learning. The evidence suggests that PLA-type schemes can be successful in encouraging individuals to take up learning opportunities that they might not otherwise have done through providing a direct financial incentive for the individual and also a wider "mobilising effect" on the population.

There is some evidence, however, that universal subsidies tend to favour those who have relatively high skills already and conversely do not provide sufficient incentive for low-skilled, low-income individuals to participate. On the other hand, targeted programmes bring with them the dangers of complexity and excessive bureaucracy that can be counter-productive.

A further conundrum associated with PLA-type programmes is the need to ensure that subsidized learning is of acceptable quality and/or is relevant to labour-market needs, while maximising the benefits of individual choice and avoiding the need for onerous quality assurance or inspection systems.

Finally, the PLA-type schemes that were reviewed - and evidence from other policy areas - points to the problems associated with using fiscal incentives to encourage individuals and employers to contribute to learning funds. A particular problem is the long period of time over which some programmes are likely to affect incentives, as well as their differential impact on people in different social, skill, and income groups.

In the light of the review, the suggested model for the practical implementation of personal learning accounts has the following broad features:

• universal entitlement at varying levels according to personal circumstances, skill type, and level and other relevant factors.
• a mixture of grants and loans, with a ceiling on the aggregate value of state subsidy.
• targeting to ensure participation of socio-economic groups that have a low propensity to take up learning.
• signposting to high-quality information, advice, and guidance tailored to the individual and linked to the proposed skills accounts.
• modest state contributions initially, along the lines of the Scottish £200/£500 model, with a built-in option to incorporate a broader range of state funding for learning, including apprenticeships, and potentially funding for higher education.
• initially managed under the Adult Advancement and Careers Service (AACS) framework (or equivalent institutions across the UK) with scope to create a "single learner bank", if deemed appropriate, in time.
• working within existing systems on assuring the quality of learning delivery and deterring the fraudulent use of funds.