Payment by Results – emerging variations in payment and performance models

Fiona Sheil was responsible for co-ordinating NCVO’s programme of seminars, training and advice work on public service commissioning and procurement. Fiona left NCVO in October 2013 but we have retained her blog posts for reference.

Payment by Results (PbR) covers a multitude of varied payment models and procurement pilots. While lots are listed and rolled-out by central Government, local government has been left to go it alone. No one is mapping the variety of models they’re trying – so learning and comparison isn’t occurring.

This year we’ve stumbled over a number of commissioners using PbR. As a bit of a catch up, here’s the variants in design that they’ve told us they’re using.

The overarching principle defining all PbR*

All PbR means:

  • paying for results (outcomes) that are achieved, rather than a block contract or spot-purchase of activity
  • paying at least a proportion of funding after results are evidenced – risk for failure to deliver results sits in part with the provider.

(* Except health services which presume clinical activity leads to a result – or they wouldn’t be allowed to do it – and therefore does PbR as payment for completed activity)

Binary and frequency models

Binary payment models

Payment is achieved only on absolute completion of a the target, with no gradation. It is a ‘yes/no’ against the measure of success. So partial achievement or reduction in occurrence levels doesn’t count to any payment. The ‘binary hurdle’ occurs where, for example, once you hit the agreed level of performance (ie 10% rise on last year) you then get paid.

Frequency payment models

Payments are made for percentages of target population who achieve outcomes. This is a stepped model, which is often capped – the more results you get, the more you are paid.

Different proportions of financing set at risk

Highly varied PbR proportions

From PbR accounting for nearly 100% of contracts, to PbR starting at only 10%. The most common proportion from commissioners we spoke to was 20% PbR.

Stepped PbR proportions over time

The PbR proportion grew by 10% each year before being capped.

Clawback

Upfront funding, with poor performance causing discounts at the end of the programme.

Views on this model are mixed, with some commissioners suspecting the upfront payment doesn’t drive enough incentive for providers to closely observe and manage their performance.

Providers stipulating proportion at risk

In at least one procurement providers decided what proportion of the fund was subject to PbR when they made their bids. This accounted for 25% of the possible score of the bid. Another 25% was on price and the remaining 50% on quality.

In this case, the variation of money bidders put at risk was between £45,000 and £180,000. The successful bidder put in the highest cost but also the highest proportion at risk.

Sensitive management of small services

Each service user represents a significant portion of the contract, and therefore can substantially alter the outcomes and attached payments.

Staged payments

Some commissioners stated that staged payments were essential to ensuring sufficient cashflow to providers; they cautioned that limiting cashflow encourages cash-saving behaviour over investment and improvement by providers.

Non-financial performance management measures

Minimum performance levels

Used in the Work Programme. This brings in accountability and enables the performance management and contract termination of under-performers. We haven’t yet seen any prime contracts terminated however.

Provider break-clause

Some Supporting People pilots allowed providers to pull out halfway through two-year contracts without any penalty. This allowed providers to exit if performance and finance risks were too great.

Longer contracts

Used to encourage providers to invest and also to enable external finance. Often with break clauses included.

Significant negotiation around performance

Commissioners reported letting providers off up to 50% of the PbR deficit they should have incurred through poor performance.

Adding additional incentives

Market shift

Giving best performers more business. This is employed by both commissioners and prime providers to encourage continual improvement.

Reinvestment of savings

A PbR pilot in Bradford on youth custody reinvested all saving locally, and producing results significantly higher than forecast.

Minimum spend requirements

Recommended by Toby Eccles of Social Finance, to ensure minimum service levels are met and not gamed.

Setting outcomes incentives sufficiently high

In order that providers are incentivised to take greatest reward from achieving outcomes, as opposed to squeezing budgets.

Piloting ‘shadow’ PbR to enable transition

This approach was popular in Supporting People pilots. This enables providers to alter their behaviour and models onto an outcomes basis without putting their payments at risk in the first period of transition.

Did we miss anything?

Have you got any further variants to add? Please comment below or email psdnetwork AT ncvo-vol.org.uk

 

This entry was posted in Policy, Practical support and tagged , . Bookmark the permalink.

Like this? Read more

Posts written by guests who have contributed to NCVO projects and events.

2 Responses to Payment by Results – emerging variations in payment and performance models

  1. Ian Marr says:

    This is a really helpful synopsis of a number of models around PbR that are emerging with some great insights.
    All these models though do seem to assume that the risk is essentially moved to the provider.
    The Social Impact Bond, and the Social Impact Investment Partnership which is about to be launched in Scotland, move the risk to a group of third party investors in a managed and structured way. This type of model has real potential to make a difference in ordinary communities across the UK and it might be good to add these models to the list for peoples information as they are also essentially PbR models.

  2. Pingback: Variations in Payment by Results processes | VoluntaryNews