What Securitisation Is and How Banks Transform Loans into Tradable Financial Instruments

Securitisation refers to a process in which several loans, such as mortgages and auto loans, are combined into a single security for sale to investors. Every person who purchased this particular security receives their part of the money, originating from the borrowers' debt repayments. So, instead of the bank waiting years for a loan to be repaid, the loan is removed from the bank's balance sheet and transformed into a tradable instrument.

Here's how this works. Imagine that the bank has 1,000 mortgage loans of £200,000 each, which amounts to £200 million in total. If the bank waits for 25 years for the repayments, it will receive cash. But instead of waiting for this amount, the bank sells these mortgage loans to a special purpose vehicle (SPV), which is a legal entity established specifically to hold the loans. Thus, the SPV issues securities that are secured by the borrowers' repayments and sells them to investors. The bank gets around £200 million in cash, investors receive money over time, and the SPV makes the process of securitisation tangible and not just an abstract idea.

Why Do Banks Securitise Loans?

There are a couple of reasons why banks securitise loans. The major reason for banks to sell loans this way is financial in nature. During securitisation, when a loan pool is sold by a bank, it transfers most of the credit risk to other agents, though it often retains low-quality tranches and offers implicit support to the vehicle it created; this was an important factor during the 2008 crisis. Second, securitisation helps banks to release their capital. Banks are required to hold reserve capital against the loans they issue, as a safeguard against losses. Removing loans from the balance sheet reduces the bank's capital requirement and frees up the related funds. Loans are originated and sold within the framework of the "originate to distribute" business model, which is crucial for the understanding of both the securitisation boom and its effect later.

Securitisation and CDOs

Securitisation and CDOs are closely related, but the terms should not be used as synonyms. Securitisation means turning loans into securities, while CDOs go further by pooling several of those securities, often backed by different mortgages, and splitting them again by level of risk. Thus, securitisation is a process while CDOs denote a specific product created using that process.

Information Problem of Securitisation

This turns what begins as a screening problem into a moral hazard one, because the institution most suited to evaluate a borrower's real credit risk has little motivation to do this well if it wants to sell that risk on.

The screening problem exists because investors are unable to view the actual loans; only the lending institution knows the real credit history of the borrower, as well as their income and collateral. This is what George Akerlof talked about in his theory of the market for lemons: one party in a transaction has more knowledge than the other, and the informed party can exploit this situation to sell the worst risk to the other, uninformed party. Gathering information on the borrower is rather costly and time-consuming, as well as providing it in a credible way to the investor, who resides thousands of miles away, and this is the gap that securitisation was developed to fill, but what it did was in fact take advantage of this gap instead.

Market Instruments for Closing the Gap

Three mechanisms emerged in the market to cover this gap. First, credit ratings substitute for the due diligence that the particular investor cannot perform, as a specialised agency is responsible for evaluating the quality of loans instead. Second, the pool of loans is divided into tranches, depending on the level of risk. Third, in first-loss retention, or "skin in the game," the originating firm is required to keep the most risky portion of the pool. This helps to align the firm's incentives with those of the investors who hold the rest of the pool.

The Path to 2008

By the 2000s, US home prices had risen for more than a decade and there wasn't any decline at the national level. The existence of a market based on that assumption made mortgage-backed securities appear safe almost by definition, because a borrower likely to default could in theory refinance or sell during a booming market. That's how subprime lending became possible; securitisation let lenders bundle those loans with less risky ones and spread the risk across numerous tranches rated by agencies that assumed defaults would occur independently.

However, the situation changed in 2007. Since the national housing market started declining, all defaults started causing a chain reaction because everyone had to endure the hardship of falling prices. In fact, the defaults caused by the downturn appeared to be highly correlated and systematic.

How the Mechanisms Failed in 2008

All three mechanisms collapsed. The agencies operate under the issuer-paid model, which creates a conflict of interest, a concept familiar from any course on the principal-agent problem. As housing prices fell, Moody's downgraded about 83% of the $869 billion of mortgage securities it had rated AAA in 2006. In total, the agency downgraded more than 36,000 tranches, with approximately one-third of them being rated AAA previously. The whole idea behind the mechanism became useless because of how the companies implemented it.

Regulatory Responses

The regulators focused on fixing the same weak points that had failed. From 2014, the Dodd-Frank Act has required securitisers to retain financial stakes of at least 5% in the assets they sell. They also tried to decrease the dependence of investors on credit rating agencies.

Every mechanism addressing moral hazard and adverse selection has its own version of the problem it fights against. Firstly, since every rating agency is an issuer-paid one, there is a conflict of interest at the very beginning. Secondly, in the case of the subordination structure, protection occurs only if defaults happen according to the model on which it was based. Securitisation did not create asymmetric information; it built a financial structure based on the assumption that such asymmetry would exist, and it failed under precisely the conditions it hadn't accounted for.