Banks and Venture Capitalists:
Are the New Rules Too Tough, Too Weak, or Just Right?
Our images are quite different. The venture capitalist -- a modern figure, dressed Banana Republic casual, high-tech, high-stakes -- resides in Silicon Valley. The bank lending officer -- formal dress, low-tech, limited loans -- resides in Main Street, USA.
While investing, venture capitalists and bankers encounter similar problems. Deciding which young companies are most likely to flourish can be very hard. Making sure that those companies which receive funds are using their money wisely and in the interests of investors, not those running the companies, can be even harder. In solving those problems, bankers and venture capitalists use similar tools.
Similarities aside, they differ in important ways, too. Venture capitalists make their investments in the form of equity, receiving shares of stock in the firms in which they invest, or of debt convertible into equity, while banks invest in debt, generally short-term. One important reason for this difference is that under the Glass-Steagall Act and the Bank Holding Company Act, banks have been strictly limited in their ability to invest in equity. That has changed significantly with the recent passage of the Gramm-Leach-Bliley Act ("GLB" or the "Act").[1]
How much have the limits on banks changed with GLB and the Federal Reserve's regulations implementing it? Is that change for the better?
Banks and Venture Capitalists: Handling the Challenges
As noted above, banks and venture capitalists have shaped similar tools to handle the challenges of choosing and monitoring the companies in which they invest. Both tend to specialize in companies of a particular type for which the bank or venture capitalist has particular expertise. Bank lending officers are expert in the needs and contours of their local economy. Venture capitalists specialize by industry.
Both types of intermediaries have more specific strategies which resemble each other. For instance, both use strategies of staged investments. Making investments in stages, rather than all at once, gives the receiving companies incentive to use the money well, increasing their chances of receiving later stages. Also, it allows the investor to update its information on how wise its investment choice has been and to limit its losses if the choice turns out to be unwise.
Both banks and venture capitalists frequently negotiate detailed covenants affecting many areas of governance of the companies in which they invest. These covenants allow the investors to block many decisions they find unwise or against their interests.
But banks and venture capitalists do differ, and one key difference is that venture capitalists typically make equity investments while banks make loans. Equity, as opposed to debt, is particularly well-suited for investments in high-risk start-ups in the high-technology sector. For such companies, the expected stream of future profits is uncertain and variable -- the companies may do extremely well, or they may be a complete bust. Furthermore, intangible assets such as intellectual property, reputation, and employee skill are often a high percentage of total assets.
These two features make such companies poor candidates for debt investment. On the one hand, if the new company crashes and burns, the creditor will be able to recoup little from its investment, as the lack of tangible assets leaves little to salvage from the wreckage. On the other hand, if the new company soars, the creditor does not share in the upside beyond receiving the fixed amount the company has promised to repay.
For these reasons, investment in high-risk, high-technology companies frequently takes the form of equity. Since outside shareholders, who lack a contractual claim to a specified stream of payments, are particularly vulnerable to managerial misuse of funds, venture capitalists will frequently become actively involved in the day-to-day affairs of their portfolio companies, often by putting themselves on the boards of those companies.
Why Don't Banks Take Equity Stakes?
When considering investing in such companies, why don't banks follow the path of venture capitalists and make equity investments rather than loans?
Once, at least some of them did. A century ago a few companies which combined commercial banking and investment banking dominated American corporate finance: J. P. Morgan and Company, Kuhn, Loeb, and Company, First National Bank, National City Bank, and a few others. Partners, directors, and officers in these banks held interlocking directorates in many of the largest companies in the U. S. and also held shares in those companies. In other countries, a fairly similar pattern remains in place today, with Germany and Japan as the most notable examples.
In the U. S., this era came to a close in 1933 with the Glass-Steagall Act. That act prohibited banks and their subsidiaries from owning shares in most kinds of non-financial companies and prohibited commercial banks from underwriting or selling securities.[2] Later, the Bank Holding Company Act stopped affiliates of banks within a holding company structure from owning shares in non-financial companies, with limited exceptions.[3]
It may be that the GLB has changed all that. Before looking at the new Act, though, I would ask whether there is any justification for the regulatory scheme which began in 1933.
Should Banks Be Allowed to Take Equity Stakes?
The limits on investments by banks are at least troubling, and require justification. After all, banks are a key source of funding for new businesses. Banks have much experience in identifying promising new companies and in monitoring their progress. Limiting the ability of banks to become involved in some of the most dynamic areas of new business may be a real loss to the economy. Perhaps bankers are not very knowledgeable in the high-tech area and would not do well in competition with expert venture capitalists. Then again, perhaps the limited knowledge of bankers in this area is due to the old rules limiting their ability to invest in equity, and with those rules removed, banks could become valuable players in this area.
Even after companies in which a bank has invested have been around for a while, a continuing role for banks in corporate governance may still be useful. Even established public companies are often not well run, and it is not clear that capital markets always do a good job of disciplining them. J. P. Morgan a century ago, and Japanese and German banks today, may help to monitor management. Increasingly those interested in good corporate government in the U. S. look to large institutional investors to oversee and discipline managers. If the law allowed them, banks might be able to do this, too.
Are there any good reasons for limiting the ability of banks to compete with venture capitalists and institutional investors by making equity investments? The main justification for the heavy regulation of banks is the unique source of their capital and the potent impact they can have on the economy, for good and for ill.
Banks traditionally get their funds for making investments from the deposits they hold. Depositors leave money with the bank, with the promise that they can withdraw any or all of it whenever they want. Banks do not hold this money in a vault somewhere, waiting to pay out withdrawals. They know that not everyone is going to withdraw all of their money at once -- they only need to keep a fraction of funds available to cover withdrawals. The money that they do not plan to keep in place as reserves is available for the bank to invest.
However, this makes banks vulnerable to runs. If for some reason many depositors think that a bank may be running out of money, they may all try at once to withdraw their deposits. If the amount to be withdrawn exceeds the bank's reserves, trouble ensues. Even the soundest of banks is vulnerable, as all hold only a fraction of total deposits available as reserves. If an epidemic of runs breaks out in an economy, it can have devastating consequences. Consumers who have lost their deposits cut back on spending. Businesses who can no longer get loans lay off workers. The payment system, which runs through banks, may break down. The economy can spiral downwards from there. This was a central part of the dynamic which led to the Great Depression.
Given this ever-present threat, long-term and illiquid investments are problematic for banks. If a run develops or appears an imminent possibility, the bank would like to be able to liquidate its investments to quickly increase its reserves. If the bank is holding shares in a start-up which is not publicly traded, the bank will find it very difficult to quickly recoup the funds it has invested in that company. Holding shares in a public company is less problematic, as the bank can sell its shares. However, the bank runs the risk of a loss on its investment if it must sell when shares prices are low, and if the bank holds a large fraction of a company's shares, it may find those shares hard to sell except at a low price.
This does not in itself justify governmental regulation prohibiting banks from making equity investments. If there is such a risk, why not allow bank managers to weigh that risk against the benefits of such investments? There are two main answers. First, the risk of a run can be contagious. If bad investments by one bank lead it to failure, that may scare depositors even in healthy banks and lead to a run on them. Bank managers have no incentive to take this effect into account. Second, in order to deal with the problem of runs, the U. S. in the 1930s created deposit insurance, so that even if a bank fails, depositors in that bank will be able to get back at least some of their deposits. This makes runs less likely and less damaging. However, it creates an incentive problem in the behavior of banks by encouraging them to make overly risky investments -- if the investments succeed, the bank makes greater profits, and if they fail to the point of causing bank failure, the government bears much of the cost of that failure. Governmental regulation is justified as a way of limiting such risky behavior by bank managers.
Even if persuasive, this argument does not yet justify prohibiting all forms of bank-related equity investing. First, perhaps banks should be able to invest some of their funds in equity, as long as such investments are not too large a share of their total investing, so that the risk is correspondingly limited. Second, banks have ways of limiting their exposure to equity investments and yet still using their expertise to become usefully involved in such investing. Rather than directly investing themselves, banks can set up related companies which make equity investments. These companies may be subsidiaries of the banks or affiliates owned by a bank holding company which also owns the bank itself. Investment by these related companies can then be funded by sources other than bank deposits. Moreover, the use of a separate corporation limits the liability of the bank to what it has invested. The banks can still use their investing skill to determine the investments their related companies make.
Such a structure gets around the direct problem created by bank equity investing. It does raise, potentially, an indirect problem. Suppose an affiliate of a bank has a large equity investment in a company. The bank may have an incentive to make a loan to that company which is overly risky to the bank on its own terms but which is worth it overall to the bank because it improves the value of the affiliate's equity investment. Equity investments by affiliates may, in this way, still lead to overly risky lending by banks themselves.
This concern is a major traditional justification for both limiting the ability of bank affiliates to make equity investments and for regulating transactions and relationships between banks and their affiliates. The concern may be somewhat valid and may justify some degree of regulation and oversight. However, the problem under this structure is less severe than where banks directly invest in equity. Other mechanisms, for instance a bank's reputation, exist to limit overly risky investing. Arrayed against the cost of overly risky investing is the benefit to be gained from bank involvement in venture financing. Banks have much expertise to offer in this area, and they potentially offer to start-ups a greater smorgasbord of services than can venture capitalists.
Several different arguments consider bank equity investing may harm the companies in which they invest or their competitors. One argument is that if banks are allowed to have nonfinancial subsidiaries, those subsidiaries will have an unfair advantage against their competitors. Perhaps this argument had some plausibility back in the days when many banks were local monopolies and product and service markets were more local. Today, with more national markets and many more sources of financing, this concern is uncompelling.
A slightly better argument is that if banks are given too much control over a company, they may direct actions which help creditors but hurt shareholders. Creditors tend to prefer that a company take too little risk (relative to the value-maximizing standard), while shareholders prefer that it take too much risk. Bank interests may thus diverge from that of shareholders. However, if the banks hold a significant equity stake in a company, their interests will diverge less from that of other shareholders. Indeed, banks that hold both equity and debt may have more incentive to maximize value than those who hold either equity or debt alone.
Thus, there is probably good reason for restricting the ability of banks themselves to make equity investments. However, bank subsidiaries and affiliated companies should have a freer rein. Although some regulation of bank relations with such affiliates, and loans to companies affiliates have invested in, is appropriate, such regulation should not be so heavy-handed as to preclude bank involvement in this area. Do the new rules under GLB strike an appropriate balance?
How Does GLB Change the Terrain?
As mentioned above, prior to passage of GLB, banks and bank-holding companies were quite limited in their ability to make equity investments in non-financial companies. Under 12 U.S.C. section 24 (Seventh), national banks are prohibited from investing in most equity securities. There are some exceptions, the most important of which is 15 U.S.C. section 682(b), which allows banks to acquire the stock of small business investment companies, so long as the aggregate of such stock owned by the bank does not exceed 5 percent of the bank's capital and surplus. Through this provision large banks can engage directly in fairly extensive venture capital investment, as long as they limit it to small business investment companies. The 5 percent limit responds to the safety and soundness concerns surrounding direct bank investment in stock described above, although one can argue plausibly that 5 percent is an overly low limit. GLB does not change these rules limiting the activities of banks themselves.
Traditionally, bank subsidiaries have been held to essentially the same limits as banks themselves. GLB does not really change this -- indeed, it reinforces the point.[4] The Act provides (in section 122) that five years after it becomes effective the Federal Reserve and the Treasury Department may adopt rules which allow bank subsidiaries to engage in the activities described below which GLB permits to holding company affiliates. This provision was a compromise.
The Federal Reserve and the House wanted to limit equity investing to holding company affiliates, while the Treasury and the Senate wanted to allow bank subsidiaries to engage in the same activities as affiliates. It is not clear whether there is any real policy advantage to allowing affiliates but not subsidiaries to engage in these activities. What is clear is that the Fed regulates holding company affiliates while the Treasury regulates subsidiaries of national banks. The agency positions become easier to understand once that fact is taken into account.
That brings us to the heart of the change under GLB -- holding company affiliates of banks. For a long time banks have been allowed to establish a structure such that a parent company owns both the bank and some other corporations. The Bank Holding Company Act has limited what type of companies can be affiliated with a bank through this structure. GLB allows banks that meet specified conditions to form financial holding companies, which are allowed to engage in some activities not allowed mere bank holding companies. The new 12 U.S.C. section 1843(k)(4)(H) allows financial holding company affiliates of banks to engage in "merchant banking," which is essentially venture capital-like equity investment.
Section 1843(k)(4)(H) imposes two limits on merchant banking activity. Subsection (iii) provides that such investments may only be held "for a period of time to enable the sale or disposition thereof on a reasonable basis consistent with" the merchant banking activities. Subsection (iv) provides that the bank affiliate may not "routinely manage or operate such company or entity except as may be necessary or required to obtain a reasonable return on investment upon resale or disposition." In addition, pre-existing limits on transactions between banks and affiliates, embodied in sections 23A and 23B of the Federal Reserve Act, remain in place. The issue then becomes whether these three sets of restrictions on merchant banking activity have achieved an appropriate regulatory balance.
Investment Time Limits
The wording of subsection (iii) is fuzzy: how long is long enough to enable the sale of the stock on a reasonable basis? This fuzziness could hurt investment decisions by bank-affiliated companies by creating uncertainty as to how long they may hold on to stock. Luckily, the Federal Reserve has created a regulation which provides greater clarity.
In February 2001, the Fed and the Treasury promulgated a final rule governing merchant banking investments allowed by GLB. Under the final rule, the Fed said that a bank affiliate may hold an investment for up to 10 years, and an interest in a private equity fund for up to 15 years. Companies may go beyond these periods only with the Fed's permission.[5]
The Fed's regulations provide a much more bright line rule than GLB itself. Considered in the light of standard venture capital practice, the time limits are reasonable -- a venture fund would rarely hold a portfolio company for longer than 10 years, and funds rarely last longer than 15 years. A better approach might have been to make the specific periods safe harbors, so that a company could go beyond the time period without having to get Fed approval, although it would make its position uncertain. This would provide more flexibility while still providing a clear safe ground.
A more basic question is why any limit on length of investment is required at all. GLB seems designed to respond to the venture capital model, where venture capitalists invest in new companies for a limited time; help those companies either go public, get acquired, or go bust; and leave. This model is a fine one, and it makes sense to allow bank affiliates to compete in this market. However, the experience of J.P. Morgan, Germany, and Japan suggests that banks can provide helpful services even for longstanding public companies and aid in corporate finance and governance even beyond the start-up period. Why not allow banks to play this continuing role if they choose? How does the investment time limit help safeguard against any of the legitimate concerns about bank investment in equity?
Routine Management and Operation
Subsection (iv) creates a potentially more serious ambiguity. Venture capitalists often take a hands-on approach to the governance of their portfolio companies. Monitoring and advising the managers of such companies is one of the main functions venture capitalists serve. If bank affiliates are seriously limited in how much of this they can do, they will be at a distinct disadvantage relative to venture capitalists. Even if the rules are simply uncertain rather than clearly prohibitive, this could cause bank affiliates to limit themselves or cause them to engage in costly legal consultation.
GLB itself is vague here. When does someone "routinely manage or operate" a company? When does it become necessary to routinely manage in order "to obtain a reasonable return on investment?" GLB does not answer these questions, and one can imagine a large number of possible answers.
Fortunately, the Fed's final rules provide more guidance. The rules list a variety of situations which constitute routine management, and a variety which do not.
Relationships which do not constitute routine management include:
- Bank representation on a company board;
- Covenants concerning activities outside the ordinary course of business, including the acquisition of significant assets, significant changes to the company's business, removal of executive officers, redemption of securities, and amendments to the articles and bylaws; and
- Advisory, underwriting, or consulting services.[6]
The rules also list some relationships which do constitute routine management:
- Employees of the bank affiliate serving as executive officers of the company;
- Executive officers of the bank affiliate servicing as officers or employees of the company; and
- Covenants restricting routine business decisions.[7]
The Fed's rules also provide some guidance as to when routine management may be "necessary or required to obtain a reasonable return on investment upon resale or disposition," such as when the company:
- Experiences a significant operating loss; or
- Loses senior management. [8]
Once action is taken to deal with the problem, the bank affiliate must stop its routine management. If such management lasts more than nine months, it must notify the Fed.
Given the constraints of GLB, the Fed's rules appear reasonable, though one can always quibble with details. The rules provide somewhat more clarity and certainty, although some examples of what do and do not count as routine management contain weasel words like "significant" and "routine," which re-introduce a degree of vagueness.
The bigger problem, again, lies with the GLB limitation itself. Why should bank affiliates be barred from routine management of portfolio companies? How would such management increase the risk the banks face? Is there any reason to believe banks would be more prone to make bad loans to companies in which their affiliates become involved in routine management than to companies in which their affiliates have comparable investments but are not involved in routine management? Indeed, wouldn't active involvement in routine management, where helpful to the portfolio company, actually decrease the risk of losses by improving the company's performance?
The routine management limit may arise from concern that banks may direct behavior in a way that privileges the interests of creditors over shareholders. However, as analyzed above, so long as banks have substantial interests as both creditors and shareholders, the conflict is lessened. Indeed, entities holding both debt and equity may well have a better incentive to maximize value created than those who hold just equity or just debt.
Sections 23A and 23B
Sections 23A and 23B of the Federal Reserve Act place a variety of limits on transactions between banks and their "affiliates." [9] The limited transactions include extending credit, issuing guarantees, purchasing assets, accepting securities as collateral, and investing in securities. The total value of such transactions with any one affiliate cannot exceed 10 percent of the bank's capital. The total of such transactions with all affiliates combined cannot exceed 20 percent of the bank's capital. Transactions must be secured 100 percent or more by collateral, and banks cannot buy low-quality assets from affiliates. Banks must deal with affiliates on terms "that are substantially the same, or at least as favorable to such bank . . . as those prevailing at the time for comparable transactions with or involving other non-affiliated companies." [10]
These limitations directly address the most legitimate worries raised by bank-related companies making equity investments. The percentage limits help assure that even if banks have incentive to make dubious deals with companies in which they have an equity stake, those deals will not commit too much of the bank's capital. The collateral requirement limits the risk posed by a transaction with a portfolio company. The requirement that transactions be substantially the same or at least as favorable to the bank as transactions with non-affiliates, while somewhat vague, does give bank regulators a tool for policing transactions with affiliates.
An important question, then, is when do these rules apply to transactions with the portfolio companies of a merchant-banking affiliate of a bank The rules apply to any company "that is controlled by the company that controls the member bank." [11] Under section 23A, a company has been deemed to have control over another company if it has the power, "directly or indirectly, or acting through one or more other persons" to vote 25 percent or more of any class of voting securities of that other company. [12] It would seem that if the merchant-banking affiliate owns at least 25 percent of the voting shares of any class of securities of a portfolio company, that portfolio company becomes an affiliate, and transactions with it are covered by the rules of sections 23A and 23B.
In addition, GLB added a new provision for merchant banking portfolio companies, creating the rebuttable presumption that if a company owns 15 percent or more of a the equity capital of a portfolio company, it controls that company. [13]
This definition of an "affiliate" probably goes as far as is needed. If the investment of a bank affiliate in a portfolio company is under 15 percent of the equity capital of that company and under 25 percent of the voting power of all classes of voting securities, then it is unlikely that the bank will have a perverse incentive to make bad loans to that company.
Still Too Tough, But an Improvement
For decades, American law has strictly limited the role banks and their affiliates can play in financing and governing new companies, by limiting the ability of banks to make equity investments. There are decent reasons for some limits on bank equity financing, but not for the limits which have prevailed in the U. S. since the 1930s.
As long as bank deposits are federally insured, there is a good argument for capping direct equity financing by banks themselves. Since Glass-Steagall the law has done so, with a 5 percent exemption, and GLB does not change this. It might be desirable to raise the rather low limit and extend it beyond small business investment companies, but some such cap is defensible.
The argument for limiting equity investments by bank subsidiaries or affiliates is less strong. Such investments may create incentives to make some bad loans, but the limits on transactions with affiliates imposed by section 23A and 23B go most if not all of the way to regulating those incentives. Little if any additional regulation is needed beyond this.
Even after GLB, regulation goes too far. Bank subsidiaries are still treated differently than holding company affiliates -- this may change after 5 years of the Act being effective, but given the Fed's ability to block such regulatory change, I wouldn't hold my breath. GLB's limits on holding periods and routine management of portfolio companies are also unnecessary. Still, GLB has gone a long way toward leveling the playing field and allowing banks to compete with venture capitalists.
The remaining excessive limits may impede some bank entry into the field but not enough to have a terribly strong effect. One hopes that as we gain experience with bank involvement in merchant banking, Congress and the Fed will become willing to go even further.
It's too soon to tell how much of a difference more bank involvement in venture capital financing will make. The law does give us a better chance to see what they can do.
Brett McDonnell is an associate professor of law at the University of Minnesota Law School. He specializes in the fields of corporate law, securities regulation, and law and economics.
[return to top]
Footnotes
Note: As a matter of editorial policy, consecutive citations to sources have been eliminated. Readers may assume references are to the immediately preceding cited source until the context indicates otherwise. (Click on the footnote numbers to return to the relevant point in the article's text.)
[1] Pub. L. 106-102, 113 Stat. 1338 (1999).
[2] 12 U.S.C. § 24.
[3] 12 U.S.C. § 1843(a), http://www4.law.cornell.edu/uscode/12/1843.html.
[4] 12 U.S.C. § 24a(a)(2)(A), http://www4.law.cornell.edu/uscode/12/24a.html.
[5] 12 C.F.R. §§ 225.172, 225.173, http://www.access.gpo.gov/nara/cfr/waisidx_01/12cfr225_01.html.
[6] 12 C.F.R. § 225.171(d). http://www.access.gpo.gov/nara/cfr/waisidx_01/12cfr225_01.html
[7] 12 C.F.R. § 225.171(a). http://www.access.gpo.gov/nara/cfr/waisidx_01/12cfr225_01.html
[8] 12 C.F.R. § 225.171(e). http://www.access.gpo.gov/nara/cfr/waisidx_01/12cfr225_01.html
[9] 12 U.S.C. § 371c, http://www4.law.cornell.edu/uscode/12/371.html.
[10] 12 U.S.C. § 371c-1, http://www4.law.cornell.edu/uscode/12/371.html.
[11] 12 U.S.C. § 371c(b)(1)(A), http://www4.law.cornell.edu/uscode/12/371.html.
[12] 12 U.S.C. § 371c(b)(3)(A), http://www4.law.cornell.edu/uscode/12/371.html.
[13] 12 U.S.C. § 371c(b)(11), http://www4.law.cornell.edu/uscode/12/371.html.
[return to top]
|
