An interesting income multiple point arose when we designed our 25-year customised fixed-rate mortgage.We were looking to make this product as attractive as possible to ensure that the consumer appetite for long-term fixes was fully tested. It occurred to us that long-term rate stability could justify a higher-income multiple. So we looked first at the average salary in the UK for the previous 10 years, which was 22,253, according to the Office of National Statistics. We then applied an average income multiple of 3.3 to this salary and calculated the monthly payment arising from the resultant loan at the average existing borrower rate for the same period, which was 7.21 per cent, according to MoneyFacts. This calculation showed that average borrowers in the circumstances described paid out 23.8 per cent of their gross monthly salary on mortgage payments. We then took the minimum salary that we were proposing for the new 25-year customised fixed, namely 30,000, and applied to that figure the same rate of average income growth over the next 10 years as had applied over the preceding 10 years. Using an income multiple of five, the highest available on the new product, we discovered that, at a 5.95 per cent fixed rate, borrowers would pay out, on average, 24.7 per cent of their gross monthly salary. In other words, if they borrowed five times income under our new product at a fixed rate of 5.95 per cent, they would pay, as a percentage of their monthly salary, the same that a borrower on average income would have paid out over the last 10 years on a 3.3 times income loan. We therefore launched this product with a maximum five times multiple – the exact multiple varied according to the fixed-rate content. It is easy to see how this calculation can be applied to the mortgage market more generally. Sections of the market are worried about misdeclaration of income, that someone might be effectively applying for a higher-multiple of income than the rules allow. But our calculation shows that, even if that is the case, it may not be as important as previously thought. The payment equivalent of five times income is what borrowers at the average variable rate have been paying out over the last 10 years anyway. Later, the issue of loan-to-income came under the spotlight when the BBC programme Panorama highlighted the use of self-certification in the mortgage market. The programme argued that abuse of this product was widespread and was causing house prices to rise disproportionately. The programme featured mainly one lender that was not always undertaking quality assurance checks. Mystery shoppers had been able, with hidden cameras, to show that a few individuals on this lender’s branch staff, and some intermediaries, were seeking to persuade applicants to massage income to obtain a higher loan than the one they were entitled using the traditional income-multiple approach. It was unusual for a lender to undertake no checks at all. Most lenders would have picked up a problem by way of an affordability model, the manual sense check or the credit profile. The house price connection was even more extraordinary, given that the growth in house prices was clearly related to a series of consecutive rate reductions, growing real incomes and a shortage of household supply versus the demand for new house- hold formation. Because it was TV, however, it led to all sorts of comment by the written press and to a thematic review by the FSA. The regulator found, as the industry knew it would, that the use of self-certification mortgages was not widespread as a percentage of total lending. It was a niche product with restricted LTV and loaded rates. Only those needing to use it actually did. Moreover, arrears’ levels on self-certification mortgages were not too different to mainstream, clearly disproving the idea that people were taking on unaffordable debt. Even if people were mis-stating their income, and there was precious little statistically sound data to prove this, they were not doing so in a way that materially affected affordability. The programme’s premise, that verifying income in a paper-based way and rigidly applying income multiples, was the only safe way to underwrite was simply out of date. Even now, more than a year later, the debate about self-certification rumbles on, sometimes with some truly daft comments attached. Changes in employment patterns mean that more people are on fixed-term contracts at work. Many have a more significant proportion of their earnings related to performance then ever before. Some have income from several sources, additional to their main job. All these borrowers are prime candidates for a self-certification product, and those which seek to prevent such employed applicants from being able to take out a mortgage are being irresponsible and deeply unfair. More people are self-employed now. Lenders are perfectly able to assess the credit quality of such applicants and to predict affordability and propensity to pay. The idea that the delays and extra expense involved in verifying income in the old-fashioned way is going to help anyone is terrifying in its naivety. The programme also sought to blur the lines between fast-track and self-certified. This introduced a broader canvas on which the programme could paint its picture. Unlike the single lender which had disproportionately attracted the programme’s attention, most lenders fast-tracked certain types of mainstream loan. They do this because making customers wait around for non-value-adding paper references was an economic cost to all parties. The information available from credit bureaux, combined with fraud databases and statistical models, provided a safer, more reliable and efficient process for verifying income. Fast-track evolved to reflect this. It was a perfectly sensible development applied to mainstream lending and is a process for better underwriting, not a self-certification look-alike. Self-certification is different. For a start, it is a separate product selected by the customer, normally because of volatile or hard to prove income, in contrast to fast-track, which is an underwriting process selected by the lender for its mainstream lending. Self-certification loans typically carry a higher rate and a lower initial LTV requirement. They share a characteristic with fast-track in that there is no income verification by way of paper reference, but there are key differences. There is no evidence to support widespread misuse or consumer detriment arising from the existence of self-certification or fast-track products. There will be consumer detriment, however, if these products become the subject of prescriptive rules based on theories and fantasies far removed from the serious business of predictive underwriting.