The Hidden Costs and Underpinnings of Debt Market Liquidity

Even as rabble rousers rail against financiers, the powers that be prize the breadth and liquidity of financial markets. Flash traders are investigated for unsettling stock markets and violators of securities laws receive jail sentences on par with violent criminals. The Federal Reserve has spent trillions with the avowed aim of pumping up the prices of traded securities, while expressing little more than the pious hope that this largesse might spill over into old-fashioned, illiquid loans.

In my article, The Hidden Costs and Underpinnings of Debt Market Liquidity, I offer a skeptical view of pro-liquidity policies. A good financial system may be vital for a thriving economy, but what warrants favoring liquid over illiquid claims? Yes, the United States—the world’s leading economic power since 1914—has exceptionally broad and liquid financial markets. But, “industry led and finance followed.” The transformation of the US from agrarian society to industrial powerhouse occurred before the smoothly functioning stock and bond markets became indispensable stars of American capitalism. The NYSE even shut down for nearly six months in 1914 without paralyzing the economy. Britain, the birthplace of the Industrial Revolution, actually lost its economic leadership, even as financial markets flourished in the City of London. Meanwhile, in spite of defeats in two World Wars, Germany’s economy and industry surpassed Britain’s, powered by businesses with illiquid stocks and banks making illiquid loans.

Economic theories that extol “complete” markets where people can trade all risks and claims at low cost are themselves incomplete: they leave out the “on-the-spot” knowledge sacrificed to sustain liquidity. Liquidity isn’t manna that rains down from the sky. Liquid markets, where buyers place unconditional bids for goods or claims offered by competing sellers, requires standardization of what is sold and how. Standardization in turn limits consideration of case-specific details—or makes them irrelevant.

Standardized terms for trading many physical commodities aren’t onerous for the typical producer or user. Buyers of copper for instance care about purity, not where and when the metal was mined. Exchanges can therefore easily sustain a liquid market in copper by specifying purity and some delivery terms. And customized transactions can piggy-back off transactions on an exchange: for instance, copper of lower than standard purity can be sold at negotiated discount to its exchange traded price.

However commodities whose buyers usually care about the same few attributes are exceptional. Desirable goods commonly combine several valued attributes: it isn’t just the size of a house or the horsepower of a car that matters to buyers. Moreover, goods belonging to the same general category often comprise distinctive combinations that hold a different appeal to different buyers. Proximity of a residential property to a high school can be highly attractive to some but a nuisance to others. And purchasers with idiosyncratic preferences usually won’t buy distinctive goods sight unseen. Houses cannot be sold by the square foot the way copper is sold by the ton. Rather, homebuyers examine properties to assess whether they match what they are looking for. Scrutinizing distinctive attributes also helps decide how much to bid: because houses are not interchangeable, the prices of similar properties are just a starting point.

In financial markets, as with tangible goods, the value of detailed case-specific information is different for different claims. Buyers of U.S government bonds expect little information beyond their maturity and coupon payments: investors’ assessments of risks and returns are based on data and forecasts about the overall economy. And, as with physical commodities, when buyers do not value case specific details, liquid markets do not require burdensome restrictions.

In other kinds of claims, purchasers care a lot about case-specific details. Venture capitalists (VCs) spend considerable effort researching startup businesses they might invest in. The research helps VCs assess the risks and returns and whether a startup matches the VCs’ expertise and portfolios. Detailed information also helps negotiate terms and prices. And, as with residential properties, the high value placed on detailed information, hinders sight-unseen trading. Investors would not normally buy shares in a startup without access to the confidential information that startups provide only to credible VCs, under non-disclosure agreements.

Then there are the in-between cases. For instance a bank can extend credit to a business by making a loan or it can purchase the businesses’ publicly-traded bonds. The bank can secure confidential information from the borrower if it makes a loan but not if it purchases a bond. Purchasing an easily tradable bond however provides the well-known benefits of low-cost diversification and the option of raising cash quickly to meet unexpected needs. And historically banks and other lenders have been willing to forgo access to confidential details mainly for large-blue chip borrowers whose creditworthiness could be reliably assessed from public data. (Limitations on the on-going information that can be disclosed to bondholders also impels differences in covenants and other contractual terms such as renewal of the credit through evergreen provisions).

In the last thirty years however the scope of liquid markets in financial claims has dramatically increased. The exceptional breadth and depth of US stock exchanges has become commonplace, while US markets have grown by trading “securitized” claims on bundles of loans, particularly mortgages, that lenders once held to maturity. Advocates hail the shift as a Schumpetarian advance spurred by innovations in information technology and finance, although they acknowledge that making more claims liquid can increase the misalignment of incentives.

I advance three propositions to critique this favorable assessment.

First, the expansion of liquid financial markets has entailed suppressing detailed case-specific information. The liquidity of stock markets has been undergirded for instance by tough insider trading rules to ensure that stocks are bought without knowledge of confidential details about the issuer’s plans and prospects. Likewise the securitization mortgages and other small loans has been undergirded by mechanistic lending practices that rely on abridged or condensed information rather than detailed information about borrowers.

Second, suppressing case-specific information imposes costs (over and beyond increased opportunities to shirk, lie or steal) that can offset the benefits of liquidity. Insider trading rules that sustain the liquidity of stocks for instance impair governance by discouraging investors from serving on company boards. Likewise using parsimonious models to screen loan applications can lead to the unwarranted extension of credit to bad borrowers and the denial of credit to good borrowers. Mechanistic securitization practices and the liquidity they help create also leads to a hazardous conjoining of risks in the banking system. Credit exposures are more heterogeneous in traditional lending where banks make loans that match their expertise.

Third I claim that regulation, rather than autonomous advances in financial and information technologies, has spurred the growth of liquid markets. The once-exceptional breadth and depth of US equity markets was the result of its unusually robust securities rules. As these rules spread, so did US style stock liquidity. Similarly small loans have been securitized to an exceptional degree in the US because of the strong bias of US housing policies, securities laws and banking rules in favor of mechanistic liquefaction.

Without such a policy bias, technological advances might have had dramatically different effects. The internet and e-commerce have revolutionized the matching of buyers and sellers of differentiated goods and services ranging from books to concert tickets to restaurant reservations. Real estate brokers routinely post pictures, maps and numerous other details of properties on their websites that help buyers quickly zero in on the ones that they would seriously consider purchasing and helps sellers reach far-flung buyers. Banks too could have similarly harnessed technology to improve their use of rich data for selecting borrowers and monitoring loans.

Rebottling the genie is difficult however. Traditional practices encouraged financiers to specialize in particular domains—say lending to restaurateurs or property developers—so that they could effectively secure and use detailed information. The declining use of detailed information had made the matching of issuers of claims and financiers with the right expertise superfluous. And to the extent the expertise is acquired through practice, even if policies favoring liquid markets are reversed, relationship-based finance will not quickly return.

The full paper is available for download here.