Government ownership of banks - the Norwegian experience
Deputy Governor Jarle Bergo
Norges Bank Conference on Banking Crisis Resolution - Theory and Policy, Oslo 17 June 2005
As it gradually became clear that the banking crisis in Norway in the late 1980s and early 1990s was a systemic one, closing and liquidating failing banks was not generally considered a viable option. The reasons why: the risk of a severe credit crunch in an already deep recession and a fear of collapse in the financial system, have already been pointed out by Kristin Gulbrandsen in her presentation yesterday morning.
However, failing banks would need to satisfy the capital requirements to continue their operations. Fortunately, in the run-up to the crisis it had been envisaged that government involvement might be necessary in the provision of capital and an institutional framework with a with a special purpose agency, the Government Bank Insurance Fund, GBIF, was established to handle such eventualities. The alternative of issuing guarantees was considered, but dismissed. I will come back to the problem of guarantees towards the end of my presentation.
In general, before the government injected capital into crisis-stricken banks, serious efforts were made to attract private investors and not only among the existing shareholders. However, these efforts did not succeed, perhaps not so surprisingly. Experience shows that in times of recession and high uncertainty, investors are extremely reluctant to take on risk. Under these circumstances, the government was the only source available to support the financial system. In a way, the government became the "owner of last resort". It was, however, the intention that the banks would be privatized as soon as conditions allowed.
The government had no ownership share in private banks prior to the crisis. The only financial institutions owned by the government before the crisis were special purpose "banks" such as institutions for educational loans and subsidized housing loans, and the Postal Bank. Thus, the injection of government capital into the private banks and subsequent ownership of them was a novelty and a challenge, even if government take-overs have been a fairly common way of dealing with banking crises in other countries. In fact, a study published in the OECD Financial Market Trends shows that this method was used in 13 of 18 banking crises worldwide.1
Government ownership of major banks raises several issues. First, I will address some of the governance issues:
- how to cope with moral hazard aspects
- how to ensure that banks return to profitability,
- how to avoid micromanagement of the banks, and
- what about the level playing field, i.a. the competitive position vis-a-vis banks that had not failed.
Next, I will discuss the exit strategy that could be applied. Thirdly, what potential problems might long term government ownership in the banking sector imply? Fourthly, is government takeover a resolution method to be recommended in other systemic banking crises?
Finally I will very briefly compare the Norwegian resolution with some features of the bridge bank operations used in the US.
1. Governance issues
The government provided capital to the crisis-stricken Norwegian banks subject to strict conditions. These included:
- the management and board of directors of the bank would be replaced
- the existing share capital would be written down to cover losses to the fullest extent possible
- the banks would not be overcapitalized
- the bank must commit itself to reduce operating costs and downsize some of its activities
- measures were taken to restrain growth in the bank's total assets.
The first two aimed at taking control of the bank's operations and containing moral hazard.
The three latter points were important not only in ensuring that the banks returned to profitable operation. They also served to prevent the banks receiving government capital from obtaining a competitive advantage over rival banks that did not receive this kind of support.
Although the members of the banks' governing bodies were replaced, the governance structure of the banks was maintained. In addition, the ownership was handled by the special purpose agency that had been set up: the GBIF. The GBIF was also the formal owner. Thus, the ownership was handled at arm's length from the government.
This set-up, maintaining the governance structure of the bank with management, board of directors, supervisory board, and shareholders' general meeting; and the government ownership executed by a special agency would have made it difficult for the political authorities to micromanage the banks - had that temptation arisen. Neither did the governance structure allow the GBIF to interfere in the bank's day-today business operations.
The crisis-stricken banks where the government had injected capital were, for the first two years after the crisis, required to report regularly and comprehensively to the GBIF on their compliance with the conditions set and their progress towards restoring profitability.
After the crisis, the financial performance of the government-owned banks was in fact not very different from the performance of banks not rescued by the government. The two major banks that were rescued had, after the crisis, pre-tax profits relative to total assets that were similar to that of other banks.
2. The reprivatization or exit strategy
No deadline had been set regarding the reprivatization of the crisis-stricken banks. This gave the GBIF some leeway in determining the strategy for selling its shares. The selling of shares could in principle wait until the economy was in relatively good shape and stock markets were not generally depressed. Furthermore, the government's shares could be sold into the market gradually. Thus, one avoided flooding the market with bank stocks and driving the stock price down. That might have been the result had the government been legally committed to selling its shares by a certain date.
In fact, the government achieved a reasonable return on the capital it invested. By year-end 2001, the government's investments showed a net positive present value, implying that the government had received a positive return in excess of a risk-adjusted interest rate.2 Such a return, of course, was not at all expected during the height of the crisis when most of these investments were made. When the special purpose agency, the GBIF, was set up and funded, its capital was regarded as expenses in the fiscal accounts, not as an investment.
3. Potential problems with sustained government ownership of banks
The Norwegian government over some years sold all of its shares in the number two and three banks, but chose to retain, to the present time, a blocking minority in the largest bank, now DnB NOR. With its current holdings in DnB NOR, the government owns some 20 per cent of the total private banking sector. The main motivation for the government to hold on to this share, was to guard against a foreign takeover of the only major bank that still has its head office in Norway.
Even if crisis-induced government ownership of banks worked quite successfully in Norway, we should recognize that such ownership poses challenges and potential problems.
As I mentioned, the government delieberately chose an institutional set-up that prevented it from micromanaging the banks. Under other circumstances, a commitment to selling the shares by a given date might have served as an alternative to reduce such a temptation. Potential micromanagement of the banks for political reasons is not, however, the only problem with the government becoming a major owner in an otherwise private banking sector. Let me mention a few of the others:
First: A potential conflict exists with the government's role as regulator and its responsibility for the supervision of the financial markets.
Second: How should the government act if a bank in which it has a substantial ownership interest fails? To reduce the first problem, the Norwegian government, after the winding-up of the GBIF and its companion the Bank Investment Fund (SBIF), transferred the management of the state ownership in DnB-NOR away from the Ministry of Finance, which is responsible for regulating the financial markets. Instead, the ownership is handled by the Ministry of Industry. One cannot of course, completely rule out that in future the political authorities might be tempted to try to interfere in the bank's day-to-day business for political purposes. In Norway, I find such a scenario very unlikely. Furthermore, the governance structure does not give a minority owner easy control over the management of the bank.
However, the second problem remains. Would the government in such a situation as easily require the value of the shares to be written down to cover the bank's losses? Politically, it might be embarrassing to admit to failure in a bank where the government itself was a large owner. Furthermore, how should the government credibly maintain a level playing field if both a partly government-owned bank and purely privately held banks encountered serious problems?
These are difficult questions, and I shall not pretend to have an answer to them.
4. A role model?
Is government takeover a role model for the resolution of banking crises? As I mentioned at the beginning of my presentation, it has been a relatively commonly applied resolution measure.
If a crisis becomes systemic and the economy is in a recession, closing and liquidating major banks is probably not a good solution. As experience has shown, under such circumstances private investors may not be willing to invest new capital in the troubled banks. Then, might it not instead be a solution for the government simply to issue a guarantee covering all the troubled banks' debt? In that way, some would argue, one would avoid the problems of government bank ownership that I mentioned earlier.
Yes, one may escape those problems, but at the expense of creating different but potentially as difficult problems. As was pointed out in the previous session where the use of blanket guarantees was discussed, such guarantees can easily create moral hazard problems, particularly in the case of banks that are already close to failing. The bank shareholders have all the upside, but their downside is limited to the remaining value of the bank's capital. Beyond that, the taxpayers have all the downside risk. Therefore, managers of financially insolvent banks may be tempted to use the government guarantee to "gamble for resurrection" by taking on highly risky projects with high yields if they succeed. Such a bank manager will not worry about an equally large downside risk, since that would be covered by the government through the guarantee. Thus, the guarantee could actually deepen the banking crisis.
To avoid moral hazard problems, a government issuing a guarantee to an almost failed bank would need to seek measures to curb the bank's activity in risk-taking. The government would then effectively run much of the bank. In my mind, a government take-over as owner is "cleaner" and more transparent. Furthermore, once a bank is recapitalized to the standard required levels, managers will not have any strong incentive to "gamble for resurrection".
If you feel reasonably confident that the governance and other potential pitfalls stemming from government ownership can be avoided then it seems clear to me that government capital injection is preferable to government guarantees.
To repeat the main requirements:
If the government injects new capital into a crisis-stricken bank, it is important that the value of the existing shares are written down as far as necessary to cover the losses. Otherwise, the government would implicitly be using taxpayers' money to subsidize shareholders of a failed or failing bank and would give rise to serious moral hazard problems.
For a government to successfully intervene in a systemic crisis by recapitalizing, and thus effectively nationalizing a large bank, it must have credibility as a "professional owner". By that I mean an owner that does not use its dominating position for purposes other than making sure the bank recovers by conducting financially sound bank business. Such an ideal temporary owner would not need a specific date at which it should be forced to sell its bank shares in the market.
As already mentioned, no explicit guarantees were given in the resolution of the Norwegian crisis, either with respect to specific banks or to the whole banking system. However, it is possible to conceive of situations and circumstances where a first-best solution is not available. Then guarantees in some form might - and I stress might - possibly be a second-best solution. The current global restructuring in the banking industry poses some challenges to either solution. Several large banks operating internationally have their head offices in relatively small countries. If such a bank was to fail and cause a systemic crisis in its home country and possibly in host countries through its branches, it is far from obvious that the government of the host country would have the fiscal capacity to rescue the bank. We may thus experience banks that are "too big to be rescued". That, however, is a topic for the next session.
5. A "bridge bank" operation?
As alluded to earlier, it was never the intent of the government to remain the majority owner of the banks it had rescued. The first sale of the government's shares started two and a half years after the peak of the crisis. In that sense, the government takeover in Norway may have some similarities to the so-called bridge bank model that Mr. Reidhill from the FDIC will discuss in the next presentation. Nevertheless, there were some important differences between the Norwegian rescue operation and the bridge bank model now followed by the FDIC:
- In Norway, the charters of the old failed banks were never withdrawn prior to government capital infusion.
- A bridge bank cannot be used as a vehicle for the government to remain in control of a bank.
Maintaining the charters of the failed banks implied that a haircut of creditors' or depositors' claims on the bank was not possible. However, in the Norwegian case any large-scale haircut of creditors was ruled out, at least once the crisis appeared to be systemic. In fact, the only time that creditors lost money was in the case of one small bank that was closed before the peak of the crisis. Now that of course is not without moral hazard concerns either.
The second point illustrates an advantage of the bridge bank model over an ordinary government takeover: it prevents the government from maintaining a long-term controlling ownership with the problems that this can give rise to later on. However, new owners have to be found quickly. In a systemic crisis that might be easier said than done.
Lumpkin, S. (2002) Experiences with the Resolution of Weak Financial Institutions in the OECD Area. Financial market Trends, (82), 107-149. OECD, Paris.
Moen, H. (2004) The present value of central government investments in and support to Norwegian banks. In Moe, Solheim and Vale (eds.), The Norwegian Banking Crisis, Norges Bank Occasional Papers No. 33.
1 See Lumpkin (2002).
2 Source Moen (2004). The interest rate used in the discounting is the current year risk-free interest rate plus a premium of 4 percentage points.