Green Gilts and the Greenium: Value for money, virtue signalling or breach of trust?

Author: Philip Bennett, Visiting Professor of Pensions Law, Durham University[1]

On 28th May, 2021 the UK Debt Management Office issued a press release which said:

“HM Treasury will publish the government’s green financing framework on Wednesday 30 June 2021 and it is planned that the inaugural green gilt issue will be launched via syndication in September 2021.”

This press release records that out of the gilt sales of £252.6 billion (cash) planned in 2021/2022, the Green Gilt component is no more than £24.5 billion (of which the minimum amount planned to be issued in this period is £15 billion).

On 30th  June, 2021 the UK Government duly published its Green Financing Framework.  This document sets out the types of expenditure that Green Gilts will finance to help meet the Government’s green objectives:

The difference between a Green Gilt and a “non-Green Gilt” is that HM Treasury as issuer will undertake to the holder of the Green Gilt that amounts raised by the issue of the Green Gilt are to be used for “green” expenditure -although not in legally binding terms:

“While it is the intention of HM Treasury to apply an amount equivalent to the proceeds of any Green Financing to Eligible Green Expenditures and to report on the Eligible Green Expenditures as described herein, there is no contractual obligation to do so.” – Green Financing Framework Section 5.  It is outside the scope of this article to discuss why a prudent person would accept a statement of intention rather than a contractual obligation.

So that HM Treasury is not marking its own homework, the issue is supported by 2 independent reports; a pre-issuance impact report provided by Carbon Trust and a second party opinion on the UK Government Green Financing Framework by Vigeo Eiris Corporation.

But does issuing Green Gilts provide value for money for the tax payer?

A Green Gilt will have higher costs of issuance than a non-Green Gilt because of the additional green advice and auditing required and the associated fees.

The Greenium to the rescue?

But green bonds issued by other sovereign issuers, such as France and Germany, have shown an interesting feature.  Those green bonds can be issued at a slightly lower yield (1-2 basis points a year) compared to the identical (as to maturity, credit risk etc but not yield) non-green bond of the same sovereign issuer.  That lower yield (or premium) has become known in the market as the “Greenium”.

For a more detailed discussion, see “Green Gilts, an overview” published by Insight Investment (Global) Limited in December 2020:

The UK Treasury’s Green Gilts structuring advisers, HSBC and J.P. Morgan, will, no doubt, have made the point that the Green Gilts are likely to represent a slightly cheaper cost of financing than if the financing had been raised through the issue of a non-Green Gilt.

Let’s do a thought experiment and assume that the HM Treasury is able raise £15 billion by issuing  20 year Green Gilts at a Greenium of 1 or2 basis points. Table 1 shows the nominal amount saved on this assumption:

Table 1

£15 billion of Green Gilts sold with a 20 year maturity


Amount of Greenium per year (basis points) Total saving (nominal amount) over the 20 year period compared to the identical (except as to yield) non-Green Gilt
1.      £15 billion 1 bp pa or £1.5 million a year £30 million  (20 x £1.5 million)
2.      £15 billion 2 bp pa or £3 million a year £60 million   (20 x £3 million)

Note: There are 100 basis points in 1%. So 25 basis points is the same as ¼%.

A freedom of information request can, no doubt, confirm, but Green Gilts, if they can be issued at a 1 or 2bp a year Greenium, should more than cover their additional green costs.

 But why would anyone want to buy a Green Gilt at a Greenium to the identical (except as to yield) non-Green Gilt?

An immediate answer may come to mind. Those investors who care about the climate change and the other issues in the Green Financing Framework are prepared to accept a slightly lower return on their investment. The fact that amounts raised are to be used for the green expenditure set out in the Green Financing Framework justifies the slightly lower return. Such an investor also has to accept that there is no contractual commitment on HM Treasury to use of proceeds for green expenditure- only a statement of intention. That may well make some investors nervous.

But let’s look at how such an answer fits with the prudent person rule that applies to trustees and other fiduciaries when investing assets belonging to others (and economic rationality).

We know that about 32% of gilts are owned, as at September 2019,  by UK pension funds and insurance companies[2].  So it is not unreasonable to expect that a portion of the £15 billion of Green Gilts would be purchased by investment managers acting on behalf of trustees of UK pension funds (whether DB or DC).

A PLSA press release on 1st July, 2021[3] said:

“Green Gilts issued by the UK government will be particularly appealing for funded defined benefit pension schemes as they are lower risk when compared to investing directly in green energy projects.”


However, a bit more analysis is needed to see whether that statement stands up to scrutiny.

Would it be a breach of trust by the pension fund trustee or its investment manager to buy Green Gilts at a Greenium?

In my opinion, clearly yes[4]

A UK pension fund trustee and its investment managers are all fiduciaries who are bound by the ‘prudent person’ rule. The same applies to EU pension funds[5].

In considering the prudent person rule and the extent to which non financial factors could be taken into account consistent with that rule, the Law Commission Report on  ‘Pension Funds and Social Investment’ published on 22nd June, 2018, said (para 1.6):

The law permits pension trustees to make investment decisions that are based on non-financial factors … provided that:

(1)  they have good reason to think that scheme members share the concern; and

(2)   there is no risk of significant financial detriment to the fund.”

This 2 stage test seems to have obtained relatively wide-spread acceptance by the Government (in relation to guidance to local authority pension funds) and by the Pensions Regulator (in its guidance to the pension funds it regulates)[6].

More recently, in the UK Supreme Court  Palestine Solidarity case decision, on 29th April, 2020 Lord Carnwath said (not central to the Supreme Court’s decision) that a local government pension fund fiduciary bound by the prudent person rule could take account of non financial factors:

“provided that doing so would not involve significant risk of financial detriment to the scheme and where they have good reason to think that scheme members would support their decision”.

“There appears now to be general acceptance that the criteria proposed by the Law Commission are lawful and appropriate.  I agree.”

This is a different formulation of the criteria (ie the 2 stage test) from the Law Commission Report (and, let us leave to one side whether general acceptance of criteria means that those criteria are legally correct).

For the moment, let us assume that each test is a correct statement of the law.

We can see immediately that Lord Carnwath’s test would not be met.  If the pension fund buys Green Gilts at a Greenium, there is clearly identifiable financial detriment- the amount of the Greenium. So, the risk is not just significant, it actually happens at the time of purchase.

In contrast, the Law Commission 2 stage test, as to its second stage (no risk of significant financial detriment), is met if the Greenium is a 1 -2 basis points a year reduction in yield. Such a reduction is a financial detriment but it is not a significant one.

So, let us now consider the first stage of the Law Commission test.

Do the trustees have good reasons to think that scheme members share their concerns on climate change and think buying Green Gilts at a Greenium for green projects is something the members would support?

Were you to ask the pension scheme members if they are in favour of the projects identified in the Green Financing Framework, you can reasonably infer that a substantial majority (perhaps close to all) would be in favour of those projects.

But, let’s rephrase the question for the members more accurately:

“ Do you agree that the pension fund should pay a Greenium to buy a Green Gilt to finance a green project which would go ahead anyway if financed by the identical (except at a higher yield) non-Green Gilt?  If yes, do you agree that the pension fund should also increase employer or member contributions to cover the Greenium”.

If you have a choice of buying 2 identical products, but one has the word “Green” written on it and costs more than the other, the economically rational answer is to buy the cheaper product.

The member’s answer may simply be “Do you think I am stupid?”.

If the first stage of the Law Commission’s test is not met, that is the end of the matter. The trustee and its investment manager would be acting in breach of their fiduciary duties if they used trust assets to pay a Greenium.

But surely the green project will only go ahead if it can be financed by the cheaper finance costs of the Greenium?

The answer to this question, on a moment’s reflection, is no.  Any value for money test on any large project will include a number of assumptions and projections (including as to the overall cost and return on the project).  To suggest that a 1, 2 or 3 basis point a year increase in the financing cost would result in the project being cancelled does not stand up to scrutiny.

This leads inevitably to the conclusion, even on the Law Commission 2 stage test, that the pension fund trustee or investment manager buying the Green Gilt at a Greenium will be doing so in breach of its fiduciary duties.

It is also common sense. Why should pension fund employers or their members pay for a cost that benefits every one. That cost should be met out of an effective carbon tax.

That said, an employer may, for branding or marketing reasons, be willing to agree to pay the Greenium to the pension fund and charge it to its marketing budget. If so, there would be, in my view, no breach of fiduciary duties.

What about DC retirement account investment options?

Including Green Gilts issued at a Greenium within the investment options reduces the amount of the member’s retirement savings corresponding to the amount of the Greenium.

But what if the member were to give fully informed consent: “I agree to reduce my retirement account by the Greenium to help finance the green project. I understand the project would go ahead anyway if there were no Greenium”.

  1. Again, the member’s answer may simply be “Do you think I am stupid?”.

We might also pause to reflect whether there is any cognitive dissonance between a DC retirement account Green Gilts at a Greenium investment option and the focus on charges, value for money and member outturns.

Just a marketing exercise in the run up to COP 26 in Glasgow?

If Green Gilts are not issued at a Greenium, the additional cost comes in as a marketing exercise by the Government signalling its green virtues borne by the taxpayer.

If Green Gilts are successfully issued at a Greenium, there is value for money to the taxpayer in the short term. However, buyers investing other peoples’ money would be doing so in breach of their fiduciary duties; so, perhaps, a boost to the fee income of class action lawyers.

NS & I is to launch a new Green Savings Bond later this year[7]. Depending on disclosure and pricing, there is the risk of a mis-selling claim to be managed.

UK fuel duty on petrol and diesel has been frozen at 57.95p per litre for 10 years.  If that duty had kept up with the increase in the Consumer Price Index ( January 2011 to January, 2021) over the 10 year period (of 19.4%) it would now be 69.18p. An alternative calculation, by Carbon Briefing in March 2020,  is an increase of 22% or 23%. Depending on the assumptions you make on price elasticity, this would, according to Carbon Briefing, have reduced UK carbon emissions by upto 5% [8].

In parallel to the Green Financing Framework the Government is taking the Environment Bill  through Parliament. Instead of at least maintaining the levels of protection that were required under the previously applicable EU legislation, it is watering down those requirements[9].

Climate change is, perhaps, the greatest current existential challenge facing not just the UK but all countries and which will have a huge adverse impact on our children and grandchildren. Much more needs to be done now, including on carbon emission taxation, rather than engaging in displacement activity.





[4] There are 3 exceptions. The employer agrees to pick up the cost of the Greenium, the active members give fully informed consent and agree to pick up the cost of the Greenium or the trust deed requires some of the pension fund assets to be invested in green bonds.

[5] See Article 19 of the IORP II Directive.

[6] The Law Commission’s conclusion on non-financial factors does not follow from the case law and would appear to be incorrect – more here:




About henry tapper

Founder of the Pension PlayPen,, partner of Stella, father of Olly . I am the Pension Plowman
This entry was posted in pensions and tagged , , , , . Bookmark the permalink.

3 Responses to Green Gilts and the Greenium: Value for money, virtue signalling or breach of trust?

  1. ConKeating says:

    an excellent and important blog

  2. Ian Neale says:

    A characteristically forensic dissection of Government marketing policy overriding inconvenient law.

    It is also interesting to see this Government proposal in the light of the DMO’s persistent decade-long refusal to issue any CPI-linked gilts to match CPI-linked pension liabilities, citing (among other claims) that it risked fragmenting the market.

  3. Pingback: “Welcome to the corridors of corruption” (Con Keating unedited) | AgeWage: Making your money work as hard as you do

Leave a Reply