Buying Second-Hand Credit: How Price Forms in the Secondary Market

Italy's distressed credit market saw roughly 22 billion euros of gross book value change hands in 2025, broadly matching the estimate Banca Ifis had published a year earlier. The composition, however, shifted. Secondary transactions fell to 47 per cent of the total, ten points below 2024, and average ticket size compressed. Smaller portfolios, more granular, mostly unsecured, with more single-name disposals.

Seen from the selling bank, those numbers read as normalisation. Seen from the buy side they read differently.

What actually changes hands

Consider a composite case built from features that recur in real transactions. A specialist fund buys a 180 million secured portfolio in 2018, around six hundred positions, collateralised mainly by residential property and second-tier industrial units in northern Italy. Purchase price: 32 per cent of face value, just under 58 million. The business plan assumes 96 million of gross recoveries over eight years, with the curve concentrated between year three and year five.

By the end of 2025 the fund has collected 54 million against the 71 the plan called for. Residual gross book value stands at 104 million, net of closed and written-off positions. The vehicle's investment period ends and the tail goes to market.

A buyer looking at that portfolio today is not buying 104 million of claims. It is buying the part the first owner could not work out, which by definition is the worst part.

Re-underwriting the collateral

Valuations attached to the residual positions are six or seven years old. Italian property has moved unevenly since: metropolitan residential values recovered, second-tier industrial units in the provinces much less, and in some inland areas the collateral is now worth less than when the originating bank booked it.

Buyers redo the exercise position by position on the top twenty or thirty exposures, which typically account for more than half the portfolio, and apply a statistical haircut to the rest. In our case, re-underwriting takes expected realisation from 104 to roughly 41 million gross.

Time is the real variable

Banca Ifis, analysing a subset of Scope-rated portfolios, reports average judicial liquidation running at 6.6 years, with 60 per cent of files open for more than five. In an investor's model those figures are not an operational footnote. They sit in the exponent.

At a 13 per cent discount rate, 41 million collected on average in year four is worth about 25 million. The same flow pushed to year six is worth 19.7. Just over five million of difference on a 104 million portfolio, generated entirely by two years of court time.

What survives the servicing stack

Servicing fees on a secured tail typically absorb 6 to 10 per cent of collections between base and success components, with legal and procedural costs on top, partly incurred and partly still to come. On 41 million gross discounting to 25, the investor's net lands around 19 to 20 million. Add the required fund margin and the bid comes out near 14 million, a shade over 13 per cent of residual face value.

Where deals break

The seller carries the same portfolio at 19 million. Not through carelessness, but because book value reflects the original business plan updated, not the curve that actually materialised. The five-million gap measures the distance between a plan built on five years of workout and a reality that consumes seven.

Hence the outcome practitioners recognise: a substantial share of secondary processes never reaches signing. Hence, too, the tendency to unbundle, selling single-name positions separately where diligence is precise and price defensible, and holding the granular tail until the procedure produces something new.

Consolidation, read from pricing

Since 2018 the top fifteen Italian servicers have become eleven, and assets under management per operator have risen from 18 to 25 billion. The standard reading is industrial maturity. From the pricing side another one appears: when margin depends increasingly on recovery speed and decreasingly on purchase discount, scale is the only remaining lever on unit processing cost.

Italy's secondary market stopped being a market in discounts to face value some time ago. What trades now is an estimate about time: how many years it takes to turn security into cash, and how much of that depends on variables neither counterparty controls. While average procedure length stays above six years, the price of a second-hand portfolio will keep saying more about the courts than about the buildings.