IPE at 20: Change pension regulation and enhance economic growth
Mutual dependence creates opportunities
Low growth is a worldwide economic problem. Western economies, in particular, suffer from the effects of secular stagnation, balance-sheet recession and/or stringent budgetary rules prohibiting extra government spending. Pension funds could help in principle, but are hindered by unhelpful regulation characterised by the dominance of financial markets.
It is argued here that pensions should be discussed in a general economic context and, with that, we deliberately distance ourselves from the predominant view of financial markets. The reason is that the mutual connection between pensions and the economy encompasses more than just financial markets as future pension payments are paid out of future national income (NI). Pension payments depend on the size of NI, the part of NI that is allocated to pensions and, therefore, on the relative income position of retired versus active people.
The idea that pension payments can be calculated from a stable distribution of returns generated by a portfolio of financial assets does not appeal. The long-term return distribution is not stable and an abnormal outcome will be amended by political forces. In the end, pensions are redistributed income in a system in which the working cohort has to be convinced to share with the non-working cohort. This intergenerational ‘contract’ is mutually beneficial and supports social cohesion.
Pension savings constitute a substantial portion of national savings. Ideally, these are deployed in order to contribute to a higher future NI, either in the form of capital income from abroad (ie, a country of rent-receiving individuals) but also from a productive use in the country itself. This article investigates these possibilities in the area of (broadly defined) infrastructure.
In many European countries, the present financial regulatory framework discourages these investments. It leads to a unilateral focus on financial markets and, by wrongly choosing the actual market interest rate as discount factor for future liabilities, imports additional risks for the pension fund participants. In the second part of this article, we therefore introduce a discount factor with a direct link to the economy. But, first we delve deeper into the slow growth in western economies, which creates a direct threat to future pensions.
Since the great recession – following the financial crisis of 2008 – western economic growth has been under pressure. The Netherlands has experienced a slight recovery, but problems remain internationally. Government and household debts have not been sufficiently reduced, still acting as a brake on public and private spending. Moreover, the balance sheets of banks are in disequilibrium, especially in southern Europe, reducing the room to finance small and mid-sized companies.
Cyclical disturbances of macroeconomic equilibrium can be amended, but this is less so for structural causes of slower growth, such as slower growth of labour supply and labour productivity.
Moreover, entrepreneurial investors are confronted with uncertainties in the area of globalisation, energy transition and climate change. Such uncertainty calls for caution in the form of a high desired return on investments and a focus on short-term projects. Research and development and investments in fundamental innovations come second. Low interest rates are not sufficient to turn the tide. More has to be done.
Productive use of pension savings
We are experiencing a period of slow economic growth. Opinions on the causes differ, see ‘Slow growth’ for a short summary.
There seems to be agreement with respect to the desire to raise investments in infrastructure, energy transformation and education. Figure 1 from the IMF World Economic Outlook 2014 (see ‘The importance of infrastructure investments’) indicates that real public investments in the advanced economies declined from 4.25% of GDP in 1970 to 3% in 2011.
Why? Are the revenues expected from public investments that low, or does the problem lie elsewhere? Literature on the expected effects of more public investments is abundant. The box on infrastructure provides a summary. In the present low-growth environment, and considering the worldwide lack of good infrastructure, it is probable that crowding-out of private investments by additional public investments will not take place; indeed, the opposite is more probable. The conclusion, therefore, is that higher public investments, especially in present circumstances, are called for from an economic point of view in that they enforce the economic structure.
The problem is broader than just insufficient infrastructure; scientific developments in IT, nanotechnology, robotisation, fuel technology and so on need more seed and development capital. For mature economies confronted with zero or negative labour supply growth, such investments are a welcome addition to the low growth potential of the economy. As in the past, (quasi-) government initiatives can be more than helpful or even necessary1 , as even large corporates find it difficult to work with longer time horizons and with greater failure risk.
The bottleneck lies elsewhere, in financing. Political arrangements on public budgets within the EU are to be blamed to a large extent and, apart from those, individual governments’ borrowing capacity is limited, especially for smaller countries.
This situation calls for bringing public investments in infrastructure to the private sector. This is possible by creating an infrastructure fund2 . This creates leeway within government budgets to invest more in education, another determinant of long-term growth3 . The argument above may have made it clear that we see an important role for pension funds in financing and managing such a public-private fund.
Here we simply outline the basic tenets of an infrastructure investment fund – details are for later. It is a private investment corporation financed by equity and debt. The government can participate in its equity but the crucial objective is participation by pension funds and other institutional investors. They provide equity, receiving an adequate return, as that helps to keep the fund’s execution at a distance from government and political interference.
Next to equity, long-term debt is issued to institutional investors, especially pension funds. The capital providers bear the credit risk of its investments without government guarantees. The expected return on these loans can be derived from the pricing of long-term infrastructural loans in the market place and consists, in principle, of a risk-free interest rate plus a credit-risk spread and an illiquidity premium. Figure 2 shows that these can be substantial.
Investments by the fund are to be assessed case by case, where each investment is evaluated on its own economic merits. Return requirements are important. We value a market-related return benchmark; this provides discipline and avoids the crowding-out of bank credit, although this long-term financing is not attractive for banks.
“ We see an important role for pension funds in financing and managing such a public-private fund”
We stress the design of the fund as a private sector entity. The fund bears the credit risk and the governance is such that political interference is minimised. The focus of the fund could be either national or Europe-wide. For pension funds, this set-up is attractive. In the final section of this article we advocate a discount rate that replaces the inefficient financial market outcome by a connection with the economic circulation in defining the rate as structural nominal GDP growth. It is possible to have a direct link between the fund’s debt and this ‘economic’ discount rate when the fund swaps interest rates with the government. Pension funds then have the advantage that an integral part of their income is not influenced by financial market volatility4 . Both pension funds and the government will benefit. In a recent publication, the Bank of England states that this is a way of risk-sharing between government and private sector which secures the long-term durability of government financing.
In the present situation, with rock-bottom rates, is it attractive for the government to enter such a swap? Yes. Nominal GDP growth is also low and gyrates along long-term government rates plus spreads for credit risk and illiquidity risk. So, even in these extraordinary times, this swap seems attractive for the government, and that is without taking into account the advantages of the overall proposal.
This leaves intact the possibility to demand a return on investment equal to nominal GDP growth as the pay-off of many infrastructure projects have a positive correlation with that growth.
Finally, it is easy to see that it would be advantageous when more countries simultaneously join in these efforts. Leakage of effective demand would be smaller and domestic pension funds would be able to take stakes in each other’s projects, enhancing diversification.
An economic discount rate
In a collective pension system with intergenerational transfers, the discount rate plays an important role. Current rules in most countries are based on the fictitious scenario of a short-term liquidation of the fund, using a second fictitious scenario in which financial markets are able to value accrued pension liabilities. Moreover, in countries like the Netherlands, it is assumed that these liabilities are nominal and risk free. This risk assumption is unrealistic and is not in accordance with pension practice. It is a political assumption, interfering with intergenerational distribution5.
Whatever the reason and whatever the specific assumptions with regard to the appropriate risk-free rate, the choice of discount rate for valuation of liabilities makes the fund vulnerable to volatility in the financial markets. It is ironic that the design of the collective pension system, covering financial risks for old age has become the opposite because of this choice. Participants are confronted with volatility and periodic derailments of financial markets and, moreover, the framework contributes to pro-cyclical movements in contributions and pension payments.
A second objection is its overriding influence on investment policy. Pension funds’ balance sheet management is core. The right-hand side is dominated by its exposure to interest rate risk. For this risk, extra buffers need to be created, unless this risk is mitigated one way or the other, for instance by compensating rate exposure on the left side of the balance sheet. As their buffers are depleted by the regulation itself and low interest rates boosting liabilities, pension funds have to hedge, leading to a concentrated investment portfolio of long-dated government bonds.
In recent years, with continually declining rates, this was beneficial for funds, but in the future these bonds will be a disaster. Balance sheet risks are hedged at the present nominal pension level, so inflation will kill the purchasing power of this nominal amount and, with a reviving economy, pensions will lag behind. In fact, pension funds without sufficient buffers to take extra risks find themselves in a poverty trap.
The real issue is whether or not there is another system for determining the discount rate that will include a fair treatment of generations. This rate must be prudent, that is, not higher than future returns to be expected from a well-diversified investment portfolio. Next to that, the rate should follow changes in the economic environment. Lastly, the rate should be a stability anchor in the pension fund’s financial management and should have a neutral effect on investment policy. Taking into consideration these properties and following our preference for an economic approach on pensions, our proposal is to fix the discount rate at the nominal growth of GDP.
This has clear advantages. Changes in economic growth lead to changes in the calculated financial positions of pension funds and will induce changes in pension payments along this road. Pension ambitions follow general welfare, be it exact or at a certain distance. The choice for nominal GDP growth is prudent. The average long-term interest rate is, over the long run, more or less equal to this growth, a bit lower in the Netherlands, and somewhat higher in the US. Equity returns, on average, are higher than this growth figure. Structural GDP growth does not show the random walk of financial markets and, therefore, can serve as an stability anchor.
We have attempted to assess the relation between pensions and economic growth from an economic perspective instead of using financial market paradigms. Direct investments by pension funds in the real economy are beneficial. As pensions eventually need to be paid from economic output, it is a logical consequence to base the discount rate directly on the central economic factor – GDP growth. That fits the macroeconomic function of pension funds as a provider of risk capital for economic growth, which contributes directly to the real component of GDP. The so-called security sought-after in the present regulatory framework is only obtained by investing in AA or higher-rated government bonds, which is disastrous for future pensions. Replacement by more real investments will strengthen the economic structure, which is the real securitisation of future pension payments.
The importance of infrastructure investments
In its World Economic Outlook, 2014, the IMF passionately advocates increasing infrastructure investments to stimulate economic growth. Governments in both developed and emerging countries should lead the way because massive external effects are to be expected and, almost by definition, these effects do not provide sufficient stimulus for private investments, focused as they are on micro profits. Necessity is twofold: stimulating present slow growth and avoiding real bottlenecks in the (near) future. These investments create a short-term demand-pull effect and a longer-term capacity effect. The first is more than necessary as monetary impulses can only influence growth to a limited extent and effective demand should be stimulated more directly, especially when present overcapacity goes along with low interest rate levels. In the longer run, better infrastructure benefits growth, which in turn crowds in private investments as indicated empirically. Moreover, proof is delivered that debt ratios decline as a consequence of induced higher growth.
Jean Frijns is a former CIO of ABP. Theo van der Klundert is an academic at Tilburg University and Anton van Nunen is a former director of strategic pension management at Syntrus Achmea.
1 For an illustrative exposition that it is government and not entrepreneurial initiative that creates breakthroughs, see M. Mazzucato (2014). The Entrepreneurial State, New York.
2 This proposal is not new. See, for example, the proposals in the Netherlands by Kremers for a National Fund for Economic Development. Of course, this initiative could easily be connected to the so-called Juncker plan proposed by the EU.
3 The total size of investments channelled through these private infrastructure finds is substantial. The size of an infrastructure fund, geared towards the Dutch economy, depends on the degree to which a rise in the public investments ratio and a parallel shift to this fund can be realised. The present investment ratio is 3.5% of GDP; if this can be raised to 4.5%, half of it by the fund, an annual amount of €10-15bn is feasible.
4 This advantage only holds when indeed liabilities are discounted by a rate equal to nominal GDP growth. Opposite, in the present regulatory framework there is a strong tendency to increase the exposure to long-term market interest rates. See: ‘An economic discount rate’.
5 See also P. Duffhues and A. van Nunen (June 2009). A better approach to solvency, Investment & Pensions Europe. They state that the choice for the risk-free rate as discount factor is both arbitrary and economically unsound.