The essence of secondary mortgage market development is the separation of the origination, funding, holding and servicing of mortgage loans. In traditional lending, for example by building societies in the UK, each of these functions is carried out by the provider. Secondary mortgage markets do not require securitisation, though this is the ultimate refinement of secondary mortgage market development and capital market integration. The key characteristic of securitisation is that when a pool of mortgages is packaged for sale to investors then the funding of mortgage debt is moved off balance sheet and is separated from the origination and servicing of that debt.
The key feature of securitisation from the investor's point of view is that they now face the risk of early prepayment or default on the debt. Thus prepayment and default behaviour will be relevant to the valuation of mortgage-backed securities. We have already seen how secondary mortgage market investors are explicitly or implicitly insured against default risk in the US. Indeed the lack of such insurance in the UK and Europe has been noted as a major inhibitor to secondary mortgage development (Jaffee & Renaud 1995). However, prepayment remains an important source of risk to cash flows to mortgage securities, even in the US.
Mortgage securitisation has been argued to have many benefits, including the noted lower cost of raising funds and thus increased efficiency in both the cost minimisation sense, and in terms of intermediation efficiency. Potential benefits include the reduction in interest rate costs for the borrower, increased credit availability and reduced credit rationing, improved liquidity for lenders along with greater degrees of geographical diversification. Later chapters will raise these issues again, for example, how far has securitisation facilitated speedier mortgage market adjustment, reducing some forms of credit rationing? Securitisation has raised other significant research questions. What are the costs and benefits for borrowers and lenders? Do mortgage rates now react more quickly to interest rate changes? Has securitisation increased the degree of integration between the market for housing finance and other capital markets, such as the bond and share markets? How do different mortgage contract designs impact upon MBS valuation and how do the dictates of securitisa-tion effect contractual features?
The growth in the secondary mortgage market in the US has provided a spur to research into mortgage market economics. The development of the securitised mortgage market should lead to the integration of the mortgage market with other capital markets; there is a stream of research that considers this question (Goebel & Ma 1993; Allen et al. 1999). The enhanced liquidity and credit controls that follow from securitisation should also lead to benefits for borrowers, including lower mortgage interest rates (see Black et al. 1981; Hendershott & Shilling 1989; and for a different view see Todd 2001). Jaffee & Renaud (1995) note how the benefits of securitisa-tion will be shared among mortgage borrowers, capital market investors and the lending and securitising agencies. Given a competitive lending market, and the relatively more elastic demand for securities by investors the benefits of securitisation are considered to be higher for borrowers (Jaffee & Renaud 1995). However, securitisation might lead to mortgage rates reflecting current interest rate changes much more quickly.
An example of how the nature of securitisation can effect loan contracts are loans which exceed the limit for conforming loans suitable for securi-tisation in the US, termed 'jumbo' mortgages. Jumbo loans have a positive rate differential over conforming loans (Hendershott & Shilling 1989; Cotterman & Pearce 1996) which might be attributable to the greater liquidity afforded to conforming mortgages through purchase by the GSEs (Cotterman & Pearce 1996). However, other explanations have been advanced for this interest rate differential, including differences in the price volatility of the houses that provide collateral for the loans (Ambrose 2001). The volatility of underlying assets effects the contract rate on the mortgage, a matter to be further explored in Chapters 9 and 10 of this book where we examine mortgage valuation and pricing.
Here we see the importance of variations in microeconomic factors (in this case, characteristics of the property) for mortgage valuation and a way that securitisation has effected the pricing of loan contracts.
So exactly how quantitatively important is the securitised mortgage market? This market is most fully developed in the US. Figure 1.1 shows the rapid growth in the mortgage-backed securities issued by the US federal home loan agencies. During 2001 the issue of mortgage-backed securities by the major US agencies reached a peak of $1,092.6 billion. This was a feature of a fluctuating market, for example, there was an issue of $482.4 billion in 2000 having fallen from $685.2 in 1999.12 The new issues of mortgage debt depend upon changes in the prime market for mortgage finance and the varying figures emphasise this link. The total amount of outstanding mortgage-backed securities in the US was $3,041.9 billion as of 30 June 2002, indicating a phenomenally large market.13
The Canadian mortgage market is argued to be undergoing a dramatic, if not revolutionary, change in funding and loan provision. Traditionally mortgage lending has been based upon a few large financial institutions. Mortgage rates in Canada, as in the United Kingdom, do not always respond to changes in market interest rates, and adjustments when they arrived were often large. However, increased securitisation and off balance
100050001980 1985 1990 1995 2000 Year
Figure 1.1 Outstanding volume of agency backed mortgage-backed securities (US$ billions).
Source: The Bond Market Association.
sheet financing is creating a more rapid response in mortgage rate adjust-ment.14 Securitisation has also developed in the Australian mortgage market being led by government house building programmes in the 1980s, which used securitisation as a source of finance. Non-deposit taking institutions entered the Australian mortgage market during the 1990s taking up to 20% of new home loans in 1995,15 thus adding further impetus to growth via competition from the retail banking sector.16 Securitisation is a truly international phenomenon.
European mortgage lenders predominantly use mortgage bonds or retail savings deposits to finance mortgage loans. However, the securitised mortgage market has shown quite rapid recent growth in some European countries. For example, France has a growth of 226.47% (in US dollar terms) of issues of MBS securities between the second quarter of 2001 and the second quarter of 2002. The equivalent figure for the UK was 18.24%, and for the whole of Europe a growth rate of 14.92%.17 The European market is of particular interest, since attempts at integrating mortgage markets can be evaluated and lessons drawn for the possibility of truly global housing finance systems. For example, extensive securitisa-tion should lead to the equalisation of mortgage rates across countries.
Stephens (2000) notes that there has been little evidence of convergence of mortgage markets in Europe and that cross-border selling has not been very successful. The main difficulty appears to be that lenders in some countries have access to 'privileged sources of funds'. It is an interesting, but at the moment, speculative, point as to whether increased securitisa-tion, and the competitive pressures which underpin it, will erode barriers to convergence on both a European and worldwide scale. Of course, similarity is not the same as convergence and while barriers to cross-border selling may remain, some similarities might develop. Generally however, we are looking at a market of international significance. The maturing home finance market is playing an increasingly important, if not yet complete, role within international capital markets.
There is a danger that securitisation, or the possibility of its increased use, is seen as a panacea for removing inefficiencies in housing finance systems, and housing markets. Indeed, there is a theoretical argument that in a perfectly competitive economy with no information problems, packaging mortgages for investors is unnecessary. Securitisation only makes sense if there are information or liquidity problems which it overcomes. There may be problems with securitised mortgage markets if primary mortgage markets are not adequately developed. For example, there needs to be the legal and real estate underpinnings necessary for primary market development. Evolving housing finance systems might be damaged by too early exposure to international capital markets, and foreign competition that hampers domestic lending and banking institutions. Globalised capital markets also increase the susceptibility of national housing finance systems to international disturbances.
The UK securitised mortgage market is not yet as fully developed as that in the US, and there are still some impediments to the expansion of this market in Europe, for example, lack of state guarantees as in the US, and varied legal and regulatory frameworks (Coles 2001). However, many of the research questions and findings that the growth in the MBS market has encouraged are applicable to understanding mortgage market issues in the UK, Europe and many developing economies with emerging and growing mortgage finance markets. For example, the fair pricing of redemption penalties to offset prepayment risk (see Skinner 1999), or the valuation of the various components of mortgage contracts (Pereira et al. 2002, 2003). Thus, this book takes the view that mortgage market economics, as currently conceived and researched, is of international importance. However, this is not a 'one size fits all' philosophy, and, in fact, a major argument of this book is the importance of meeting and analysing the variety of needs and preferences, both within and across evolving housing financing systems.
Was this article helpful?