The guarantee needs fine-tuning
International leaders have been meeting to find a way of coordinating national responses to the current financial crisis. The alternative appears to be a race to the bottom of national safe harbours created by extensive government guarantees of deposits, bank debt, nationalisations and other assorted measures.
In a crisis, introducing broad guarantees may be a sensible strategy, if the government is assured of several things. The first is that in the absence of a run by depositors or an investor boycott of bank debt, the banks are solvent, sound and secure. The second is that an imminent run threatens the solvency of the banks. Then, if a guarantee prevents the run or solves the boycott (as it should), the issuance of a guarantee has no direct monetary consequences for the government.
Even in those circumstances, however, there are good and bad guarantee policy approaches to choose from. In particular, guaranteeing outstanding bank debt is poor policy. Guaranteeing new issues of bank debt, for a fee, may be good policy.
For Australia, this is an important issue because of the heavy dependence of our banks on international capital market debt funding. The Prime Minister’s announcement on Sunday that ADI deposits and (for a fee) debt securities would be guaranteed illustrates the importance and severity of the international crisis.
At the risk of oversimplification, Australian banks raise debt funds internationally by issuing fixed rate bonds, floating rate bonds and short term commercial paper. Fixed rate bonds are issued at a fixed coupon rate involving a fixed credit spread over the government bond rate at the time of issue. The cost of borrowing, the interest bill, is thus fixed over the life of the bond.
Floating rate bonds are issued as a fixed credit spread over some variable indicator rate such as Libor, and movements in Libor over the life of the bond affect the interest rate amount paid by the bank. Commercial paper is rolled over as it matures and the new interest rate reflects current market rates and credit spreads required by investors.
In the current crisis, nervous investors have increased the credit spreads they require on bank debt. Consequently a fixed rate bond initially issued at par, at say, 50 basis points above the government rate, may now trade at 90 cents in the dollar, because investors have increased their required credit spread from 50 to, say, 200 basis points.
What is the implication of that for the bank which issued the bond? The egos of directors and senior management might be bruised, but there is no direct effect on the bank’s cost of those funds. That was locked in when the bond was issued.
The disturbing effect is that new issues of debt, to expand the balance sheet and replace maturing debt or rollover short term paper, will be at the higher cost of funds involving the 200 basis point credit spread. If that spread blows out further, and a bank cannot raise funds at rates which can be passed on to its loan customers, the bank’s ability to fund itself and survive is threatened.
Issuing a government guarantee over all of the bank’s debt resolves this problem. The market price of outstanding debt snaps back towards the issue price of one dollar (more or less depending on how market interest rates have moved and the credit rating of the government relative to the bank issuer’s original rating). The bank is able to issue new debt at the government bond rate.
If the bank really is solvent in the absence of a run, this looks like a free lunch. And it would be to the international holders of the outstanding bonds – at the expense of the Australian taxpayers. While there is no monetary outlay by the government, it has passed up the opportunity to sell such a valuable guarantee to those investors. Taxpayers have suffered an opportunity cost.
There is a much better way to tackle the fundamental problem, which is the cost banks face of raising new debt funds, rather than the market value of outstanding bank debt. Simply guarantee new debt issued by the bank – until such time as the crisis has receded and guarantees are no longer needed.
Of course, if the guarantee is at no cost to the bank, this would be a free kick to bank shareholders. The guarantee is worth something and should be charged for – realistically a price which is “historically” normal for the credit rating of the bank.
Even this gives a potential free kick to the bank. Suppose the guarantee is priced at one cent per dollar of bonds issued. If the bank’s outstanding bonds are trading at ninety cents, there is a strong incentive for the bank to repurchase and cancel those bonds and issue new bonds at a price of one dollar for a net gain (after paying the one cent guarantee) of nine cents.
Perhaps that arbitrage opportunity should be constrained by regulation, but it will probably be constrained through market forces driving the price of the outstanding bonds toward their par value. Hedge funds, for example, would perceive the opportunity to undertake risk arbitrage by going short the newly issued guaranteed bonds (for cash inflow of one dollar) and buying the old (non-guaranteed) bonds (for cash outflow of ninety cents). Market forces are likely (even in these troubled times) to drag the price of the outstanding debt back towards a more appropriate value.
Government guarantees of newly issued bank debt do not, unfortunately, necessarily signal to the market that these are temporary measures reflecting the known inherent strength of the banks rather than a propping up of weak potentially insolvent institutions. Hence other measures are warranted to provide such signals.
One is demonstration of bank profitability, and Australia is well placed at the moment for a coordinated policy to provide such a signal (which should precede or accompany any introduction of guarantees). Three of the major banks have just completed their financial year and are finalising their accounts.
Rather than the usual delayed, staggered, competitive, speculation inducing, individual releases of results, a government coordinated joint early release of strong profit results could have significant impact in international markets. Auditors might need to abandon their usual reticence and be willing to make conditional statements along the lines of “At this stage of our audit we have no reason to believe that bank XYZ’s profit will lie outside of $2-4 billion” to achieve that.
Desperate times call for innovative, but preferably not desperate, measures. Hopefully, the release of more detail about the announced guarantee scheme will indicate that only new debt issues are covered. And to maximize the benefits for the Australian financial system, coordinating and hastening bank profitability information announcements look to have merit.
Kevin Davis is Commonwealth Bank chair of finance, University of Melbourne, and director, Melbourne Centre for Financial Studies.
Disclaimer: the author holds bank shares and is a taxpayer.