Edition: Sep - Nov 2014


Into infrastructure

Relative to what lies ahead, private-sector investment is in its infancy. There’s huge potential for pension funds. Trustees should get a grip on where they stand.

Smit...essential guidance

Just when it’s most needed, along comes this driver for SA economic growth. Private-sector participation in the financing of public infrastructure, traditionally the preserve of the state, opens exciting opportunities across the spectrum from job creation to improved risk-adjusted returns on the portfolios of pension funds now allowed greater diversification.

But hang on a moment. Potential must relate to practicalities. There are stumbling blocks.

Old Mutual, for instance, has pointed to the shortage of technical specialists in infrastructure development and considers the targets for capital formation envisaged in the National Development Plan to be “ambitious” (TT June-Aug). Group economist Riaan le Roux notes that, although big numbers have been talked for some time, as yet there’s been no infrastructure boom and little net growth.

This has not inhibited a boom in conferences about infrastructure.  They’re well attended by enthusiastic delegates from the pension-fund industry, sometimes with smatterings of trustees. Most recently, in June, the Association of Savings & Investment SA assembled a top-rate panel on private-sector funding to accord with NDP objectives. In July the Bowman Gilfillan law firm hosted an impressive array of delegates from across the continent keen on moving the discussion from why to how pension funds invest in Africa.

It’s easy to answer why. Around the world, fiscally-constrained governments must turn increasingly to the private sector for the funding of new projects; no less in SA, as the NDP makes evident. That capital outlays are huge, with investment returns materialising over decades, makes them ideal for pension funds; at least in theory.

Moving from why to how is less easy. Although the marriage of long-term investment returns from the projects and long-term investment returns for pension funds is made in heaven, short-termism continues to pervade the thinking of fund trustees. Too often, if an asset manager fails to make it into the top quartile over a 12-month period, trustees threaten to replace him. Also, trustees don’t know where to look for big infrastructure projects and are at a loss to assess them when found.

“Pension funds will need to build up their own capacity to assess the projected viability of different projects, or find a way to cooperate or outsource this with other players,” says Nicky Prins of National Treasury. “An understanding of the underlying risks is always important.”

Especially with mid-sized pension funds, a lot of work with trustees is needed. “Much relies on consultants,” suggests Alastair Herbertson of Investec Asset Management. “They’re the advisors and have the capacity to upskill. A key step is for trustees to identify their risk appetites.”

NDP commissioner Elias Masilela points out that SA’s biggest problem is not finance but the allocation of resources. Should the private sector not become involved, even to take the lead, he warns of monopoly structures; inefficient delivery of goods and services; asset decay arising from low maintenance and poor usage efficiency, and intensified competition for stretched state finances which will necessitate higher taxes.

That’s a scenario to send the private sector scurrying for action stations. And if they don’t start scurrying – Masilela noted that the NDP does not discuss funding – there’s always the Damocles sword of prescribed assets again to threaten. He was too polite to mention it.

Essentially, the private sector means financial institutions. Significantly, these institutions embrace pension funds. Practically, these funds can either invest themselves or through mandates to their asset managers. They also have choices between a range of instruments that include development bonds, private equity and directly into specific projects.

There are differences. For instance, an investment in a government-guaranteed bond (the most orthodox and conventional component of a fund’s portfolio) will show a return that moves with the market rating of government bonds. On the other hand, direct investment in say a toll road will show a return that moves with the profit it makes.

Trustees need to navigate a welter of choices. Adré Smit, a senior policy advisor at ASISA who convened the conference panel (see box), synthesised extracts from the discussion to guide trustees on the basis of questions posed by TT. The theme of the questions assumes hypothetical trustees of a mid-sized SA pension fund wanting to consider infrastructure and to learn what they should do to go about it.

TT: Approximately how much money is required for investment in SA infrastructure over the next five years? Of this amount, are there any indications of how much might reasonably be expected from SA private-sector pension funds?

Smit: Over the past five years government has spent R1,1 trillion on public-sector infrastructure. It has planned to spend another R1,5 trillion over the next five years and has budgeted to spend R847 billion over the next three years. While government has indicated that it wants the private sector to assist in the funding of infrastructure, other than for the renewable energy Independent Power Producer (IPP) programme, it has not specifically come out and indicated where it wants the private sector to participate.

Nevertheless, ASISA is working closely with National Treasury -- together with the Banking Association of SA, the Venture Capital Association of SA and organised labour -- on a task team for private-sector funding of infrastructure. This initiative is seen as leading edge within an international context and has been showcased at numerous international conferences.

The SA savings industry has some R7,7 trillion of assets under management. This gives it capacity to participate substantially in the infrastructure space, but the R7,7 trillion needs to be contextualised.

At present the savings industry has some R1,5 trillion invested in listed government and public-sector debt, an amount which can be viewed as investment into infrastructure. This is already significant because it represents 19% of the assets under management. In terms of direct investment into infrastructure, so far there’s been only some R70 billion or less than 1% of assets. These figures are not out of line with global experience.

Infrastructure investments are mainly driven by state-owned enterprises such as Transnet, Eskom and Sanral. They tend to raise money by issuing traditional bonds, so depriving investors of direct exposure to the underlying assets. Might it not be better for funds to invest directly in the underlying assets and, if so, what would be the appropriate funding vehicles?

Direct investment into infrastructure assets is seen as a good match for the liabilities of the long-term savings industry. They have a long tenure and their returns are linked to inflation or gdp growth. A further benefit is that their returns are not as volatile as those of listed instruments. It therefore makes a lot of sense to invest directly into the underlying assets.

The constraints, however, are that there is a shortage of bankable assets on the one hand, and on the other hand there is a shortage of skills to analyse these assets within the savings industry. This is why, so far, there is only R70 billion invested. Industry is responding to these challenges by skilling up its resources.

In terms of the appropriate funding vehicles, a project bond that has line-of-sight to the cashflows of a project would be appropriate for those with the necessary skills to analyse the asset. For others, there is scope for the creation of special-purpose vehicles that contain a number of projects from which a range of instruments can be created. These could include senior debt, subordinated debt, mezzanine debt and equity.

The revised Regulation 28 took effect in July 2011. Has it facilitated pension-fund investment in infrastructure? Has it had much impact?

Within the revised Reg 28, capacity has been created for the investment of up to 15% of a portfolio into unlisted debt instruments and a further 10% into private equity (plus a further 2,5% into ‘other assets’) that would cater for direct investment into infrastructure.

At this stage there is no evidence that this capacity has resulted in additional investment into infrastructure. But the industry is gearing up to use these limits.

The long-term nature of infrastructure investments is eminently suited to the long-term nature of pension-fund investments. But how would the funds then meet their short-term liquidity requirements to continuously pay out benefits? Any advice on how funds should take care of this aspect in making their decisions on strategic asset allocations?

Infrastructure investments are typically long term and illiquid. As such, only a proportion of a fund’s portfolio should allocated to them. A portfolio should always have sufficient liquid assets to meet its short-term requirements. These would be outside its investments into infrastructure.

Getting away from the general principles, where and how would a particular pension fund find a list of possible investments suited to it from which it could select; for instance, to choose between say toll roads and renewable energy?

A diversified portfolio of infrastructure investments is not a current reality. In the few instances where this has happened, it has taken 10 to 15 years to build up such a portfolio and been done by institutions that have developed their skills over this sort of time period.

Where there are bankable projects, they are highly concentrated; specifically those projects created by the IPP programme. As the pipeline of bankable projects grows, managers will develop their expertise around specific asset types and will then be able to provide a wider choice.

Where do trustees access information on the projected viability – the risk/reward factors – for the investments available? Can trustees place reliance on such projections, or would prudence dictate that pension funds only seek investments where there are government guarantees?

For sound economic cases there is no need for government guarantees. There are opportunities for investment, but to invest in them you’d need the requisite skills. An asset manager is required to do its own analysis so it is not able to rely solely on third-party projections.

There is no reason why a pension fund should not invest in these assets if its asset manager has those skills.

What would be the bigger risks, distinct from the asset classes in conventional portfolios? By the same token, could one anticipate bigger rewards?

Infrastructure assets are typically owned by governments that have the ability to impact on the cashflows the assets generate through the determination of tariffs. The biggest risk in this asset class would be inconsistency in government policy.

An example would be a toll road where government has agreed on a tariff formula over the life of the transaction, where this formula has been the basis for the transaction to be priced. If this formula is tampered with negatively, the investor will not get the return anticipated.

A further consequence is that, if this does happen, it will be difficult for government in the future to sell similar transactions for private-sector investment. There would typically be an expectation of bigger rewards because the project is not guaranteed by government, but that premium would be determined by the confidence that the market has of consistent government policy over the life of the transaction.

Within an infrastructure project there is a lifecycle of risks, each requiring a different set of funding and/or funders. There is the building or greenfield phase, where the risks are higher because no cash-flows are yet being generated, so demanding a higher return. It’s followed by the mature or brownfield phase when cashflows are being generated, so the risks are lower as would be the return.

What about private equity?

The pension fund will have to rely on the skills of its asset manager to entertain instruments like private equity.

Are their blockages to investment by pension funds? If so, what are they and how are they to be overcome?

Reg 28 does not currently create a blockage to investment into infrastructure. The time may come when there is a need to amend the regulations specifically to cater for direct investment into infrastructure.

A perceived blockage is that pension-fund mandates to their asset managers do not cater for investment into infrastructure.

To accommodate these investments, how should trustees redraft their investment policy statements and mandates to asset managers?

The first step would be for pension-fund trustees to buy into the benefits to be derived from investing into infrastructure. Once done, the next step would be to amend their mandates to asset managers specifically to invest into infrastructure in an incremental way up to a certain percentage of their portfolio.

Cognisance also needs to be taken of the long-term and illiquid nature of infrastructure investments in the mandating of an asset manager. An appropriate period must be allowed to realise the return on the assets.

Should SA pension funds be looking only to invest in SA infrastructure, or should they also be looking abroad and particularly at the rest of Africa where opportunities abound?

As it stands, Reg 28 has a specific allowance for a further 5% of a portfolio that can be invested into Africa over and above the 25% that can be invested offshore into all asset classes. Purely on the basis of credit rating, SA would be a less risky jurisdiction than the rest of Africa. The experience of the managers that have invested into Africa is that those governments have been pretty accommodative in terms of engaging with the private sector.

Prudently, one would expect the bulk of a pension fund’s infrastructure investment to be made in SA.


The panel discussion was on infrastructure and the NDP. Put to the panel was: ‘The NDP believes there is a critical role for the private sector in financing infrastructure development. What does this mean for the financial-services industry?’

Panelists were:

  • Nicky Prins, chief director at National Treasury;
  • Elias Masilela, an NDP commissioner;
  • Mike Peo, head of infrastructure, energy and telecommunication at Nedbank Capital;
  • Louis Strydom, principal for business development at DBSA;
  • Tom Plaistowe, an executive for infrastructure and development at Old Mutual;
  • Alastair Herbertson, an investment specialist at Investec Asset Management;
  • Johan Marnewick, an executive of the Liberty Group.