PROPOSALS for a credit rating override in the Pension Protection Fund (PPF) levy framework could significantly increase levies for big businesses, reports sister title Professional Pensions.
The government-run ‘lifeboat fund’, which companies with defined benefit (DB) pension schemes must contribute to in order to protect member benefits in the event of insolvency, has proposed using credit rating agency (CRA) scores where available, instead of its bespoke model for calculating levies.
CRA scores would be used to allocated a risk-based levy band, which reflects the size of the business and the likelihood of it being unable to fund its DB liabilities.
The PPF said its current model, which was developed with Experian, was less effective at predicting insolvencies at big companies than for its population as a whole.
It said CRA scores should be more accurate than its own model for companies with assets of more than £500m because the agencies could conduct a more rigorous analysis.
But Towers Watson senior consultant Matt Roscoe said an override would “move the goalposts” for the biggest firms because levies would increase in most cases where an override would be applied.
He said: “Some scores will improve, but the majority will be worse off. In most cases we have looked at the company would move down one or two bands, but we have one client who would move from band one to band eight. That would mean a 900% increase in its levy.”
As only the largest employers are rated by one or more of the agencies, the numbers affected would be relatively small. Roscoe says 2.5% of levy-payers will be directly affected, but warned there could be an indirect impact on up to 15% of companies.
“Because these will be some of the biggest schemes, an override could affect a big proportion of the levy,” he added.
In its response to the PPF consultation, Towers Watson also called for the PPF to consider increasing the number of bands it divides employers into.
The lifeboat fund is actually considering cutting the number of bands from ten to eight, because it says it cannot usefully distinguish between all of the bands.
But Roscoe said the consultation should have looked at whether the 20% of sponsors in the lowest risk band could be split up.
He said: “Their argument is that there isn’t enough evidence. Clearly that is the case because there are few insolvencies in that top bracket.
“But for the not for profits they put a lot of effort in and went outside the data set to differentiate. Why can’t they do the same for band one?”
He also said there was no justification for using 12-month average ratings to calculate levies under the new model, which will be much less volatile than the system it replaces.