Housing Finance Reform and the Shadow Money Supply.

• Author: Min, David

1. INTRODUCTION II. HOUSING FINANCE AND THE MONEY SUPPLY A. Fannie, Freddie, and Shadow Money Supply 1. Defining Money 2. Shadow Money Creation 3. Shadow Money and the Demand for Collateral 4. Safe Assets as Collateral 5. Fannie and Freddie Obligations as Collateral B. Private-Label Mortgage Securities and the Money Supply C. Thrift Depository Institutions and the Money Supply 1. Thrift Deposits and the Money Supply 2. Thrift Institutions and Money Creation III. HOME LOANS AS "NATURAL" COLLATERAL FOR MONEY A. Home Loans Are a Sizeable Asset Class B. Mortgages Are Good Loans Overcollateralized by Valuable Assets C. Housing Finance and Monetary Transmission D. High Costs and Long Duration of Housing IV. EFFECTS OF HOUSING FINANCE REFORM A. The Housing Finance Reform Debate B. Short-Term Effects of Housing Finance Reform 1. Substitution of Privately Created Safe Assets 2. Substitution of Foreign Safe Assets 3. Greater Rehypothecation C. Long-Term Effects of Housing Finance Reform V. IMPLICATIONS A. Financial Stability Effects B. Monetary Policy Effects C. Caution in the Pace and Structure of Housing Finance Reform VI. CONCLUSION I. INTRODUCTION

   In a previous Article, (1) I showed that government guarantees in the U.S. mortgage market have historically promoted financial stability, and argued that this stability was at least in part due to three effects of government guarantees: first, they prevent banking panics; second, they blunt procylicality; and third, they promote the origination of consumer-friendly loans that are less likely to default.

   In this Article, I propose another complementary explanation for this phenomenon: housing finance naturally produces liabilities that function as money. Thus, government guarantees of housing finance liabilities create two positive externalities promoting financial stability--they facilitate greater linkages between the Federal Reserve's traditional monetary policy levers and the actual money supply, and they crowd out private forms of money (like the "shadow money" liabilities created by the shadow banking system), which are more prone than publicly backed monetary instruments to procyclicality and instability.

   This hypothesis may have a great deal of importance, given the likelihood of major reforms to the housing finance system under the Trump Administration. In September 2008, the government-sponsored enterprises (GSEs) (2) Fannie Mae and Freddie Mac were placed into conservatorship by their primary regulator, the Federal Housing Finance Agency (FHFA), a casualty of the financial crisis that was then brewing. (3) Following their conservatorship, a wide array of commentators (including many influential legislators) have been clamoring for some form of "privatization" of the GSEs, wherein the two companies would be wound down, and the federal government would be removed from its longstanding role as the guarantor of last resort of most U.S. housing finance liabilities. (4) President Donald Trump appears to be sympathetic to this view of housing finance, as his presidential transition team was made up of several leading advocates of housing finance privatization. (5)

   Such reforms could potentially have a staggering impact on the mortgage markets in the United States, as the two enterprises provide a very large share of all residential mortgage financing. Since the 1990s, and particularly since 2008, Fannie and Freddie have been responsible for financing most American home loans. (6) Thus, any such major structural changes to U.S. housing finance will likely have enormous implications for American homeowners and home buyers, and more broadly, the American economy. Unsurprisingly, legal and policy analysis of housing finance reform has focused almost entirely on the effects this may have on the mortgage markets.

   But GSE reform would also have a major effect on global credit markets, a point that has been mostly overlooked to date. To finance their considerable mortgage finance activities, Fannie and Freddie have issued approximately $5.3 trillion in mortgage-backed securities, collateralized mortgage obligations (CMOs), and corporate debt. (7) These liabilities, which are understood to carry an implicit federal guarantee against losses that is nearly as robust as the "full faith and credit" guarantee on U.S. Treasury obligations, are an important component of the supply of so-called "safe assets," making up about one-tenth of the estimated global supply of safe assets, and about one quarter of U.S. safe assets. (8) As I discuss infra in Part I, safe assets vary greatly in form and function, but are generally thought to carry de minimis credit risk. (9) In the absence of any countervailing measures, housing finance reform is likely to have a short-term contractionary impact, possibly a very severe one, on the supply of safe assets.

   Any reduction in safe assets would in turn have important consequences for global money markets today, as safe assets, particularly government-backed safe assets like the liabilities of Fannie and Freddie, have come to play an essential role in the creation of "shadow money." Shadow money has become an important part of the overall money supply, but because it is a private form of money creation and does not carry any formal government guarantees behind it, it relies heavily on the use of collateral to help maintain its monetary attributes. (10) As the International Monetary Fund has described, collateral serves as a "lubricant" or substitute of trust in transactions between financial institutions. (11) Safe assets, which are thought to carry de minimis credit risk, are strongly preferred as collateral in shadow money transactions.

   To the extent that housing finance reform may significantly reduce the supply of government-backed safe assets, as seems likely, this may have negative ramifications for financial stability and monetary policy. As Prof. Morgan Ricks and others have argued, money creation--including shadow money creation--is inextricably tied to financial stability. (12) In the short run, we can expect to see two major effects of any housing finance reform that significantly reduces the supply of government-backed safe assets. First, there is likely to be a substitution effect as shadow money markets rely more heavily on privately created safe and foreign safe assets. Second, to the extent that there are insufficient substitutes, such a reduction in government-backed safe assets will create contractionary pressures on the shadow money supply, which has become an important part of the overall money supply today. In tandem, these two near-term effects are likely to be quite detrimental for financial stability and macroeconomic growth.

   If, as this Article argues, housing finance is intrinsically important for the money supply, then housing finance reform will also have some sizeable long-run effects as well. At the very least, we would expect to see a large increase in the stock of private money, which would greatly raise the likelihood of future financial crises. While there has been much debate over the causes of the 2007-08 financial crisis, there is little dispute that one of the key factors in the striking lack of similar crises or panics in the preceding 70 years was the prevalence of government backing for money liabilities, which effectively prevented bank runs by eliminating concerns that these money obligations would become worthless if they weren't immediately redeemed. Since private money does not carry government guarantees behind it, holders of private (or shadow) money claims must worry about run risk, and thus private money is inherently vulnerable to crises of the sort we just experienced.

   Housing finance reform then may result in both severe short-term stresses and large long-term structural deficiences in our financial infrastructure. Policy makers must begin to account for these "supply side concerns" in their deliberations over what to do with Fannie Mae and Freddie Mac. If they do not, we run a high risk of re-creating the same conditions that led to the financial crisis.

   This Article proceeds in five parts. Part I describes the historical importance of housing finance liabilities for the U.S. money supply. Part II argues that housing finance may play an important role in the money supply because home mortgages have characteristics that make housing finance uniquely well-suited for producing money liabilities. Part III describes the directionality of housing finance reform and discusses some of the likely effects of housing finance reform on safe assets and the money supply. Part IV lays out the implications of the arguments made in Parts II and III, and cautions that housing finance reform should proceed carefully in reducing the government's footprint, as this will have significant effects on the money supply, which in turn could have important and negative impacts on financial stability and monetary policy.

2. HOUSING FINANCE AND THE MONEY SUPPLY

   Housing finance reform will heavily impact global money markets because housing finance is deeply intertwined with money creation. As I describe in this Part, the obligations issued by Fannie Mae and Freddie Mac play a substantial role in the overall money supply today. These obligations are heavily relied upon as "safe" collateral in a variety of shadow banking transactions that create liabilities that serve the functions of money.

   The importance of housing finance to the overall money stock is not a phenomenon that is limited to Fannie and Freddie, but indeed has been the case as long as we have been measuring the money supply. Before Fannie and Freddie, thrifts were the dominant source of housing finance in the United States, and the deposits they used to fund their activities, much like the MBS and debt issued by Fannie and Freddie, were an important part of the money supply. Also, the brief period in which private-label MBS dominated the U.S. mortgage...