John McLaren provides a thoughtful response to George Kerevan's view of Scotland's fiscal position and his imaginative idea that an independent Scottish government could smooth the flow of oil revenues through the sale of 'covered bonds'.
I am not a finance expert but some simple arithmetic suggests to me that the covered bond idea would not offer much in the way of smoothing. If I have got this all wrong then I'd be pleased to hear what I am missing.
Kerevan's suggestion is as follows:
There is a simple solution to getting an oil fund up and running quickly. The new Scottish Sovereign Wealth Fund (SSWF) could sell what are known as "covered bonds", to capitalise future oil payments. Covered bonds are a form of mortgage, in which securities are backed ultimately by dedicated cash flows such as rents – or in this case oil revenues. The interest rate would be fixed relative to the average oil price.
The SSWF then uses the capital it has raised to buy foreign equities, bonds and property that yield a steady income. True, if oil prices drop temporarily, so does the SSWF's income. It may have to sell assets or issue new covered bonds to support existing interest payments. But the average oil price curve is upwards, so the SSWF can recoup short-term interest rate losses in later years, especially when oil is expensive. Either way, the problem of volatility is removed from the Holyrood budget.
Below I set up 3 scenarios that seek to operationalise Kerevan's proposal as I interpret it.
The main assumptions are drawn from John McLaren's article: a non-oil fiscal gap of -£8bn per year and oil revenues that vary in the range from £10bn to £5bn, with poor year scenario 3 where oil revenues drop to £2.5bn. Otherwise everything is the same across the 3 scenarios.
The SSWF issues covered bonds up to £200bn. Yes, that is what would be required to generate sufficient income from re-investment. The interest rate paid on the covered bonds is 3%. I've fixed it since that is the normal practice with such bonds but I guess it could have some relation to oil prices as Kerevan suggests, since oil revenues are supposed to back the bond. The funds are then re-invested at a higher rate of 4% - the same as that earned on the Norwegian Government Pension Fund. The covered bond pays out less because it is less risky being 'covered' by an asset, oil reserves and the tax revenue stream. The SSWF in investing in equities, bonds and properties is taking on more risk and so the return is greater.
If the two rates are fixed then the SSWF and the Scottish government gets a constant net income stream of £2bn per year. However, the government fiscal balance is still variable according to the scale of oil revenues. In Scenario A, oil revenues are £10bn a year and with the Oil Fund the fiscal balance is +£4bn. If oil revenues drop to £5bn, the fiscal balance is slightly negative at -£1bn and if the oil revenues half again to £2.5bn, the deficit worsens to -£3.5bn.
So, it would appear that Kerevan's proposal does not deal with the problem of oil revenue volatility. It also ignores the transactions costs of setting up the SSWF, issuing the covered bonds and re-investing the proceeds. Moreover, it is not clear that backing a covered bond with oil revenues would work with investors. Normally, the asset pool for covered bonds contains mortgages on property and/or public loans. Moreover, the creditor is given a direct claim on the asset, which may not be feasible since as McLaren notes the oil revenues depend on volume produced - companies may independently decide to leave the oil in the ground - the price of oil and the tax rate.
Of course if there was no non-oil fiscal deficit then the oil revenues could be invested as Norway has done in the Government Pension Fund building up over time to a significant asset pile. But to achieve that in an independent Scotland, taxes would have to rise and/or public expenditure fall to remove the non-oil deficit.
Comments