This brings us to the second issue in this section: the collateralizability of sources of financial firm value.

In a perfect market with complete information and contracts, there would be no distinction between value and collateral. The value of virtually any asset could be realized quickly in the marketplace. In this world, our definition of capital would be the market value of the firm less its liabilities.

It is important that perfect collateralizability is a more comprehensive notion than perfect liquidity, and therefore much harder to achieve. For example, suppose a publicly-traded firm sells additional shares. Prices may become temporarily depressed because it takes time to assemble buyers with incremental demand at incipiently lower prices; this is what we mean when we say liquidity is not perfect. However, something else happens when an established firm sells shares: shareholders receive a negative signal about true value. Because managers tend to sell when their private information suggests overvaluation, an announcement of an equity sale is negative information the market previously didn’t have. The result is that share prices fall. The pre-issuance share price is not achievable, and therefore the firm’s market value is not collateralizable. The larger the offering, the larger the price decline. This is true even if the shares are perfectly liquid in the usual sense.

Note that equity-issue price declines do not have to be very large to substantially reduce collateral on call. For example, suppose that an ongoing firm has investment opportunities but currently is low on internal funds. It has a market value of 1000 and needs to raise 100 to fund the opportunities. If the price decline is 5% (which is typical in information intensive industries for this size equity raise), then the market value of the firm including the additional 100 is 1050 (1000-50+100). The firm issued 100 in new shares, but increased its total value by only 50. Thus, the cost of collateralizing market equity is 50% for this firm.

The degree of collateralizability naturally differs across sources of value. The assets in place (e.g., loans) may be illiquid, but they can often be collateralized at close to fair value. This is especially true if the central bank is prepared to purchase such assets directly. But even in the absence of central bank intervention, there are liquid markets for loan pools. Evidence for this comes from CLOs, where pools of loans are sold to special-purpose financing vehicles at competitive prices. Other, less tangible assets cannot be so readily collateralized. The seller of claims to highly intangible assets, such as a firm’s growth opportunities, faces the same information and control problems as the firm that announces an equity offering. Growth opportunities cannot be sold easily, because their capture involves an intricate combination of inputs, including specific individuals, knowledge, information, relationships, etc.

Much academic work has sought to put more structure around this thinking.5 Because there is asymmetric control over and information about future profits, it is expensive to collateralize them. Firms will find it costly to raise external funds in large amounts based on such opportunities. The costs are greater if the opportunities are opaque and if managerial discretion is greater and smaller if they are transparent and contractually perfected.

To continue with the example of the bank that lends at a riskless 10% spread, the market might well value rolling the one-year note spread at full net price – 200 if the opportunity were funded internally. However, if the firm were to try to raise external equity to fund the loan, the equity would sell likely at a lower price. Information asymmetries and adverse selection make shareholders skeptical that managers see the prospects as good. Of course, the closer that the bank can come to demonstrating publicly that the loan must be rolled, the more the profits become more transparent and mechanical and the more cheaply they can be collateralized.

This logic forms the basis of theories of risk management and capital allocation, applied both to financial and nonfinancial firms.6 In these theories, firms have “internal” funds (e.g., reasonably liquid assets in place) that can be used to make investments or cover losses. Firms can gain access to external funds, but there is a cost to doing so, since that requires them to collateralize informationally-intensive profit opportunities. Additional risks that they take on must be priced according to how much additional costly collateralization will be needed.

Clearly, we cannot include the market value of all firm assets in a sensible definition of economic or regulatory capital. Transformation of the market value of the firm into internal funds involves a reduction in value. This value reduction is low if a well-capitalized firm borrows short-term funds, and it is large if a firm with little financial slack needs to raise a considerable amount of equity. Many internet companies in 2001 are an extreme example of the latter, with positive market value but access to external funds only at near-infinite costs.

What about using only next-quarter profits as opposed to the entire future stream of discounted profits? Once again, we have to ask how easily such profits can be collateralized. Clearly, if the firm has untapped debt capacity then such profits might be financed without much cost. However, if the firm is run thoughtfully, it should not have untapped debt capacity. Even in the extreme example of Microsoft above, we should probably view the excess debt capacity as dedicated insurance collateral. If that case, even Microsoft would endure important costs in collateralizing one-quarter ahead profits. Drawdown of insurance collateral depletes the firm’s financial slack.

To conclude this section: firm market value in excess of warehouse-asset market value should not be immediately counted as capital for economic and regulatory purposes. Indeed, a conservative view would not count any of the excess as capital. In fact, a conservative view would also take into account the liquidity risk of the warehouse assets, thereby further reducing capital. In individual instances, of course, such a conservative view may not be justified: a portion of the firm-warehouse market excess might be collateralizable. Unfortunately, however, there is no easy way to determine generally the degree to which collateralization is feasible.7, 8

5 For a survey of the issues and literature, see Froot, K., “Incentive Problems in Financial Contracting: Impacts on Corporate Financing, Investment, and Risk Management Policies,” in The Global Financial System: A Functional Perspective, Harvard Business School Press, Boston, 1995.

6 See Froot, K. and J. Stein, “A New Approach to Capital Budgeting for Financial Institutions,” Journal of Applied Corporate Finance, Summer 1998b, 59-69; and “Risk Management, Capital Budgeting and Capital Structure Policy for Financial Institutions: An Integrated Approach,” Journal of Financial Economics, 47, January 1998a, 55-82. Also see Froot, K., D. Scharfstein and J. Stein “Risk Management: Coordinating Corporate Investment and Financing Decisions,” Journal of Finance, 48, December 1993, 1629-1658 and “A Framework for Risk Management,” Harvard Business Review, 72, September-October 1994, 59-71.

7 We still also face the issue of illiquidity of balance-sheet assets. See Stephen Kealhofer, “Liquidity, Liquidity Crises and Bank Capital Regulation,” discussion paper, 2001 for a discussion of these issues.

8 In the above VaR simulations, we counted profits as they were received, and we assumed that in expectation, profits accrete continuously over time.