Note on Financial Stability and Supervision No. 11 - Bad loan recovery rates in 2016

The Bank of Italy released one new Note on Financial Stability and Supervision (Note No. 11).

This note updates the bad loan recovery rates estimated in Notes on Financial Stability and Supervision, No. 7, and provides a comparative analysis of the characteristics of closed bad loan positions relative to stocks outstanding at the end of 2016. The data, along with some detailed breakdowns that are commented on but not reported in this text, are available in digital format. Our analysis reached the following main conclusions.

In 2016 the overall amount of closed positions reached its highest level since 2006 (€17 billion, compared with €13 billion in 2015), mainly owing to the increase in sales on the secondary market. This value is still lower than that of newly classified bad loans, but the difference has more than halved over the last three years. The number of positions closed exceeded that of positions pened.

In 2016 the average recovery rate came to 34 per cent (35 per cent in 2015 and 34 per cent in 2014). The slight decline reflected the significant increase in the share of positions sold, whose recovery rates remain far below those obtained using standard recovery procedures. Nonetheless, the average recovery rate for positions sold on the market did increase (23 per cent compared with 20 per cent in 2015).

The speed of disposal of bad loans is still rising gradually. The share of positions closed within one year of being classified as a bad loan, which had fallen to a low of 20 per cent in 2012, rose to 38 per cent for the positions opened in 2015. As for the amounts involved, the ratio between the amount of bad loans closed each year and the stock outstanding at the start of the period, which had reached a low of 6 per cent in 2013, was consistently above 8 per cent in the last three years, reaching over 9 per cent in 2016.

The recovery rates observed recently do not appear to overestimate potential recoveries in the next few years. In fact, the age of the positions closed during the period 2006-16, for which the recovery rates have been calculated, is much higher than the average vintage of the stock of positions outstanding at the end of 2016. These closed positions also show a smaller percentage of positions secured by collateral. Their composition by counterparty sector (households, firms) and geographical distribution (north, centre, south) appears very similar to that of the outstanding.