As you probably know, the European Commission recently launched a consultation on the Green paper on the long-term financing of the European economy. The deadline for the consultation is 25 June 2013.
By sharing the draft responses to each question on Long Finance, he welcomes questions, comments and criticisms beyond EFFAS' traditional audience. Download EFFAS draft submission or read below EFFAS draft responses to the questions asked in the consultation. Please do not hesitate to leave a comment here.
We do not agree fully with the analysis presented; it is incomplete in a number of aspects. However, we do not believe that an extensive critique is productive, but will point these differences out in our other responses, wherever in our opinion they are material. Our responses are, in the interest of brevity, sketches rather than detailed implementation plans.
We feel that the following quotation from Colin Mayer serves well to give context to our responses to these questions:
"This combination of shareholder interests, contracts, reputation, regulation and state engagement underpins the structure of economies around the world. It is the basis of national and international policies of domestic and global public institutions, When and where markets fail, there is a need for more regulation and state engagement; where state organisations malfunction, the privatisation and more liberal markets are required. This economic and political consensus emerged progressively towards the end of the 20th century and is now widely accepted. However, it has serious defects. Equally misconceived is the alternative, unconventional paradigm that advocates a private sector approach to the problem –‘corporate social responsibility’, Social entrepreneurship’, and ‘stakeholder values’. These see the fundamental problem as lying with companies and markets, and a need to reposition both to broader social agenda. They seek to realign them by altering the objectives of companies and markets. Where they fail is in establishing credible criteria by which these objectives can be delivered and in ensuring an alignment of the interests of socially conscious people with the priorities of their wider communities. The failure of the conventional and unconventional paradigms is in providing a compelling description of the corporation."
This viewpoint conditions our responses and the various proposals we put forward.
We are particularly concerned that the paper is single-minded; there are competing demands for capital, notably from households (for housing), together with governments (for investment) and the private sector, over which we would have concerns about potential unintended consequences – particularly over the question of ‘crowding out’. We believe that it is difficult to discuss long-term financing without touching upon the conflict between short-termism and long-termism as outlooks. In that context there is one significant area where we believe the coverage in the analysis is inadequate and it is an area that should be expected to feed into the responses to this consultation. This concerns short-termism and its role in fostering institutional corruption. This is a strand of research emanating from the Edmund Safra Centre for Ethics at Harvard University. The seminal paper is Lawrence Lessig’s “Institutional Corruptions”. In particular, we would draw attention to Malcolm Salter’s recent paper: “Short-Termism At Its Worst: How Short-Termism Invites Corruption… and What to Do About It”. We quote from the abstract of that paper:
“The central concern is that short-termism discourages long-term investments, threatening the performance of both individual firms and the U.S. economy. I argue, in this paper, that short-termism also invites institutional corruption. Institutional corruption in the present context refers to institutionally supported behaviour that, while not necessarily unlawful, erodes public trust and undermines a company’s legitimate processes, core values, and capacity to achieve espoused goals. Institutional corruption in business typically entails gaming society’s laws and regulations, tolerating conflicts of interest, and persistently violating accepted norms of fairness, among other things.”
Though US-centric, we feel that there are many insights and lessons to be learned from these publications which may transfer to the European context and long-term investment.
We need to change our culture regarding the long-term and in particular its relationship to the short-term. All too often we still hear that clarion call of short-termism; that the long-term is simply a compounding of lots of short terms, so that the only required focus is to take care of the short term and the long-term will simply derive from it. Most often reflected in the single-minded emphasis on quarterly returns in the stock market and within the fund management industry, this attitude largely misses the point. With such an attitude, would Boeing ever have designed the Jumbo jet? So, the most important aspect of any definition of long-term financing is the definition of long-term.
As we will see below, “long-term” has a number of attributes: we first focus on the simple, temporal one.
We can start by establishing a floor in the definition: long-term must mean longer than 20 years (i.e. longer than a generation). We would suggest that there is no real need to establish a ceiling to the definition, because there is none. For example, there should be no end to the period during which motorways and bridges are maintained: and that maintenance requires long-term funding. When the US Interstate Highway System was built under President Eisenhower, the Federal Highway Act included the establishment of a Highway Trust Fund, funded by hypothecated tax receipts on petrol. Such a fund should have no end while there are still cars using the highways but, in an all too typical example of the short-termism that we deplore, the US Congress has repeatedly refused even to tie the tax rate to inflation. Any consideration of the long-term must recognise that there is much in our political system which supports short-termism. This example should highlight the value of a long-term view, and the folly of a short-term one (at least in this instance).
Certain projects of value to the commonwealth (such as hospitals and universities, or roads and bridges) have no maturity date: the long-term for them is forever. They should be funded in a similar manner, and the hypothecation of specific taxes is not the only way. Capital markets have, in the past, been happy to issue and buy undated, perpetual bond issues, and this was not confined just to government issuers (the Canadian Pacific 4% issue springs to mind). Equally some of the players mentioned above, and in particular some charities and many universities and colleges can take particularly long-term positions on the back of particularly long-term views: Balliol College in Oxford is celebrating its 750th anniversary this year (which is a full 220 years older than the world’s oldest bank still in existence). The sources of long-term funding exist, and so does the demand.
We are reminded of the presence of cathedrals and churches that grace so many European cities – the financing, so many centuries ago, of their construction raises more than a passing resemblance to current issues, and perhaps even extends to aspects of institutional organisation. The sales of corrodies (a pension taking the form of board and accommodation) were notable in this regard.
We have emphasised the extreme of what we mean by long-term, and long-term financing, and are fully aware that the supply and demand of long-term funds will be for periods shorter than forever. However, it is important that we not impose any arbitrary constraints on our views. In brief, long-term financing is financing for at least one generation, and with a possible maturity date of forever. We will discuss later some of the inter-generational issues.
But there is more to long-term financing than the issue of time. In their work, the OECD have associated long-term investment with patient, productive and engaged capital. It is clear that time alone does not define the long-term. We note that the question refers to financing rather than investment, but will draw out some important though broader distinctions.
The patient, productive and engaged attributes are, we believe, symptomatic of the long-term rather than causal or defining. For example, many observers consider mortgage finance to be concerned, inherently, with long-term investment. This is undoubtedly true in the case of new construction; there it meets the OECD productive criterion. However, the majority of mortgage finance is concerned with the purchase of components of the existing stock (previously occupied housing); this is, in the owner-occupied case, the long-term financing of housing consumption rather than housing investment . In many European countries, the consumer attitude to owner-occupied housing is that it is investment, and it is the subject of much speculation. In a consumption context, of course, this is not the case and low rather than high prices are then preferred. We recently heard a variant to this in the context of the UK NEST pensions savings, where the recent rises in stock market prices was welcomed. However, the auto-enrolment system is expected to generate some £11 billion of annual contributions when the system is fully operational; these savers should prefer to buy their investments at low, not high prices. This is also relevant in the context of arguments with respect to intergenerational inequity. We return to this issue later.
This housing finance example raises a question of purpose; in this case, in the use of the financing. We believe there is a similar issue of purpose with respect to the source of the financing. Savings can have many different motivations or sources, from the relatively short-term to the extremely long. This can be from saving for some specific near-term purpose, such as the purchase of a car or holiday, to the extremely long-term of pensions and inheritance bequests on death.
One of the widespread failings of current national statistics is that volumes of savings that are measured and published, but there is little or no discussion of, or data on, the term profile of these savings. This term profile absence is true, also, of the usage of these finances – though the recent estimates of infrastructure and climate change investment demands are a good starting point. We would like to see not just estimates of the volume of savings and investment demand, but also their term structure.
This interest in the term structure is directly relevant to the role and ability of markets and banks to perform the maturity transformation that resolves inter-temporal imbalances between supply and demand. We discuss a related concern with traded equity markets later. Liquidity is a critical issue with any maturity transformation – the ability to refinance when a deposit is called in the case of a bank, and the ability to sell a security at some future time in the case of a market. Financing which relies upon inter-temporal transformation should be expected to be more expensive than financing which does not, precisely because it is reliant upon liquidity conditions, which have a cost.
With this in mind, we should like to propose another (and complementary) method by which we may distinguish the short from the long-term; the source of liquidity from which the contract will derive its performance. Here, we would define the long-term as reliance upon the obligor of the contract as the source of the cash-flow liquidity, both before and at expiration of the contract. By contrast, we would classify reliance upon the market as the source of liquidity as short-term and speculative.
In this view, the purchase of a treasury bill and its retention until maturity would be long-term, as would the purchase of listed equity that is held solely for the collection of dividends. A life insurance policy is non-negotiable and relies upon the life company obligor for performance. A characteristic of the long-term is that income is the dominant concern in performance, which contrasts with the short-term where price behaviour is almost all-important. It is also notable that the high volatility, which characterises short-term behaviour in financial markets, converges slowly to the relatively low volatility of fundamental performance and growth when long-term horizons are considered.
It is as well to remember how low this long-term variability (or risk) has been. The diagram below is reproduced from a recent Bank of England speech. Formally, this is the standard deviation of rolling 30-year periods of annual GDP growth.
It is clear that there are many shades of grey in this liquidity distinction. The speculations of high frequency trading are clearly short-term and the fifty year zero coupon gilt held to maturity is clearly long-term. However, in between, there are many differing degrees of dependency upon obligor versus market sources of liquidity. There are degrees of implicit preference that arise from this source of liquidity approach. The source of liquidity for an industrial enterprise may be entirely independent of the financial markets, sourced entirely in the sales of its products. The source of liquidity for contracts with financial institutions is usually predominantly financial markets. However, the non-negotiability of many consumer financial contracts serves to reduce the dependence of these institutions upon the markets, and incidentally exposes mark to market accounting standards as being inappropriate for these institutions. Here, the risk issue is not whether the instrument can be negotiated or sold to others, but whether the contract can be negotiated with the institution itself, such as deposits called from a bank.
It should also be recognised that financial institutions may add value in manners which are not replicable by an individual. The maturity transformation of banks might be replicated by an individual operating in traded securities; the bank, however, has an advantage with respect to the information it has with respect to its loan customers. However, there are many forms of collective organisation, insurance companies, DB pension schemes and mutual funds, which offer risk-pooling and risk-sharing advantages that are not replicable by an individual. This is the source of ‘financial depth’, and one of its consequences is that, if efficient, it will lower the aggregate need for savings. This suggests that savings targets should be specific to the economy and its financial infrastructure. It also suggests that there should be diversity in the form of ownership of these institutions.
The concept of replication figures prominently in market consistent accounting. Liabilities are replicated by traded assets. This rather begs the question: if an insurance or pension contract can be replicated by traded assets, why do these institutions, insurance companies and pension funds exist at all? Reliance upon the contract can be seen as ‘patient’ investment. We have some concerns over the ‘engaged’ aspect. We would prefer the concept of committed. Engagement in practice often means no more that lobbying management for higher dividends and immediate performance, with a thinly veiled threat of exit by sale in the market. The long-term investor is committed, for the term of the contract in the case of complete non-negotiability. This is the case in many savings contracts. We shall return later to issues of commitment and the control rights associated with the long-term.
It should be recognised that the long-term is already effectively defined in much regulation. This is most obvious in the case of (long-term) capital gains taxes. The application of preferential rates of taxation with holding term and purpose has been widely utilised within Europe to provide incentives for specific forms of investment. It is not obvious to what extent these incentives have been successful, or whether they have been cost-effective, let alone optimal. There are also many implicit issues within other financial sector regulation, for example: the risk weights applicable to long dated bonds, or even the setting of insurance risk margins from a ‘hedged’ base position within Solvency II.
It is, of course, also possible to define the long-term in terms of business cycles. In the mid-20th century, Schumpeter and others proposed a typology of business cycles according to their periodicity, to which names have been appended:
One of the attractions of this type of approach is that it would be simple to distinguish between some of the classification issues evident in the Green Paper. This relates to the term structure of demand for capital investment by users, for example, the role of R&D (discussed later) in the fixed investment and long technological cycles. Finally, we would point out that some long-term pensions and savings entities have already declared their view of the long-term; in Canada, ten years has been chosen, while the Singapore GIC considers the long-term to be 20 years and more.
Of their nature, traditional deposit-taking commercial banks are ill-suited to execute maturity transformations for long-term periods of the type we have used to define long-term above (i.e. from 20 years to forever). We have long held that the best way to minimise overall risk in the banking system is to recognise that there are numerous types of lending, expertise in which is not fungible. Thus, trade lending is not the same as mortgage lending, and consumer credit lending is not comparable to long-term project lending. In particular, whilst we recognise that there is a very valid and constructive role for securitisation within the financial system, that role should be played mainly by the capital markets and less so by the banks (and there should be a clear distinction drawn between them).
We perceive that one of the lesser-discussed contributing issues of the recent financial crisis was the lack of heterogeneity among the business and operating models of banks. This was competition beyond competence and informational advantage. Studies by Standard & Poors indicate that the banking sector has historically provided more than 60% of infrastructure financing while pensions, insurance and asset managers have supplied slightly less than 20%. We also note that capital market financing has been small relative to bank debt – approximately one tenth.
There have, of course, been specialised long-term credit banks (particularly in Japan, where they existed within a very structured banking system), and it is not impossible to create a system which includes specialised long-term credit institution. Having specialised institutions of one kind in the banking system does raise the question as to whether or not banks should specialise in a small number of areas in which they have specialised expertise, such as shipping and project finance, or trade finance or mortgages. This question is beyond the scope of this response, but it should be noted that the creation of specialised long-term investment banks has implications for possible changes elsewhere in the banking system. The initial performance of the UK Green Investment Bank, which has already disbursed more than €700 million, suggests that specialist institutions may rapidly acquire momentum.
Long-term credit banks are encouraged to fund largely through the capital markets, with issuance of long-term bonds. With the pressures on banks to improve their capital and liquidity ratios, the prospects for increases in bank lending on the scale necessary to accommodate the projected increases in infrastructure do not appear good. It also seems unlikely that they would be able to increase the degree of longer-term debt financing in their overall capital funding on the scale necessary to maintain the current maturity transformation mismatch, let alone reduce it.
It should also be realised that the long-term investment sector will not be able to increase its capabilities in infrastructure finance to substitute for banking on the scale necessary. Its principal constraint would not necessarily be capital, though the four-fold increase required if banks do not increase their lending, would be extremely challenging. The obstacle would be human resources, the limited supply of the necessary skilled and experienced staff. The investment could most easily increase its exposure to capital market instruments.
If the ring-fencing or creation of utility banks with a retail mandate occurs and is sustained, these banks would likely operate as brokers and advisors for their clientele in capital markets; their role as conduit rather than principal. The asset management industry can be expected to create specialist retail infrastructure investment funds and distribution through the retail banking system could be expected – however, the level of demand for these is highly uncertain.
The way to best support the financing of long-term investment and the pursuit of EU policy goals (inasmuch as these involve long-term investment) through multilateral and national development banks is to encourage the latter to further develop the depth and maturities of the liabilities that they issue to finance themselves. These could include far more securities with targeted investor bases and far from the simple, homogenous, fungible instruments of benchmarks and active trading markets. This could include such things as deferred term annuities or bonds with sinking funds and active amortisations. This could also include securities or instruments offered directly to individuals, and include non-negotiable instruments.
It is arguable that the EIB is a significantly better credit than any of the member states of the EU, its shareholders. Given the likely increased demand for safe assets from the banking sector (for capital and liquidity purposes) and from the derivatives sector (for the collateral requirements of CCPs), there is more scope than ever for the advantageous issuance of securities to satisfy these demands. Here, there is a question for these shareholders of how much maturity transformation it is desirable to undertake within these institutions. It would seem that there is a role for the commercial banks as originators of infrastructure debt and as suppliers of liquidity insurance to securitisations of this debt. However, many of the characteristics of infrastructure finance, such as the construction risk, are idiosyncratic and not readily amenable to securitisation. There is also the question of the level of retention, a necessity for the management of the moral hazard inherent in the originate and distribute model of securitisation. On this retention issue, we would suggest that it should be possible to have retentions which declined with the passage of time.
Multilateral and national development banks should co-ordinate their liability management and engage in ever-closer cooperation when it comes to funding long-term infrastructural projects. One way to confirm that there would be economies (of whatever nature) to be gained from increased co-operation would be to encourage the formation of working parties between official institutions to determine where to find such economies and how to exploit them. Later, we recommend the formation of a new infrastructure financing institution. This could be organised as a bank, but equally it might be an asset manager or even insurance company – in fact, a hybrid institution would seem to be optimal.
Survey work by Standard & Poors on infrastructure projects suggest that multilateral institutions account for only 3% of total financing, smaller than governments at 9% and only marginally higher that the export credit agencies. We would suggest that the EC should,
Prior to the 2007-2008, much infrastructure financing was greatly facilitated by indemnity cover supplied by the mono-line credit insurers. In fact, these companies originated in the US municipal market, where much of the longer dated issuance was infrastructure investment related. We suggest later that a specialist infrastructure bank should be created, which we envisage would actively supply the guarantees and indemnities. On a technical note, we would suggest that these should be of subrogation form, where the bank would step in assuming coupon and principal payment responsibilities, rather than financial guarantee, where capital sums are exchanged immediately.
In our other responses, we indicate the potential for the issuance of deferred term annuities. Many investors have expressed concern over the risks of the construction phase of projects and have avoided financing this phase. We wonder if guarantees from the development banks might resolve this issue; however, we would indicate this is an area for further thought rather than an explicit recommendation on our part.
There are complementary policies and broader frameworks that would support the financing of long-term investment.
We will begin by noting that investment in the context of the Green Paper includes both research and development expenditure (R&D) and capital investment. We find it helpful to consider R&D as being concerned with the investigation of options and flexibilities available to the institution while capital investment is the exercise of those options. This difference alone suggests that different financing methods are appropriate. External debt finance is unlikely to be available for R&D simply because, in and of itself, R&D does not deliver future revenues that may be pledged, even implicitly, to creditors. We note that there is already an EC objective of raising R&D expenditures to 3%, with two thirds of this being financed by the private sector, from the (approximately) 1% of today.
There is a school of thought that, in Europe, current R&D expenditures are overly focussed upon increasing the operational productivity of existing processes and not adequately upon the types of innovation that might prove more useful in recovery from recession. If that is indeed the case, then R&D expenditures need to be refocused on areas which are less geared to proving real but short-lived productivity benefits to existing processes, towards a less certain but potentially much more productive . Foe example, according to President Obama, the US Government’s investment in the Human Genome project is expected to have a return of 140:1. In terms of the cycles mentioned earlier, this is the long Kondratiev cycle of technological change. These innovations are now frequently referred to as disruptive technologies, and can be very important in terms of employment, wages and questions of social inequality. It seems likely that these will prove important in the coming decades.
The majority of investment in the EU economies is undertaken by the corporate sector (in this response, we use the expression corporate sector to include partnerships and unincorporated bodies). As several of the later responses make clear, it is critical to remove any impediments to the corporate sector participating in investments. This applies both to the taxation treatment of financing and to procurement policies. This is particularly important in the current situation. We would also note that the post crisis fall in private investment in the EU27 was extremely substantial (€300 billion plus) and far more important than the fall in private consumption. We would also note that private investment has historically been about six times as large as government investment.
We note that the paper is concerned with external finance, while most corporate investment in fact utilises internally generated funds. We would also note that the corporate sector is currently extremely well positioned to increase investment as its cash holdings, counting only listed companies, now exceed €0.75 trillion. In Europe, 40% or more of long-term investment is financed by corporations, with government typically accounting for volumes that are in the 25%-35% range. The balance is accounted for by households. The European corporate sector’s reliance upon external finance is relatively modest –averaging in the 30%-35% range in France and Germany and slightly higher, 45%-50% in the UK.
The rate of internal generation of funds by the corporate sector can be improved upon. For example, through the increased use of book-reserve pension schemes – these can be structured and insured to offer efficient and secure pensions to employees, while reducing the dependence of the corporate sector on external (mainly bank) funding. Funded DB pensions, by contrast, constitute a material drain upon the cash resources of their corporate sponsors in the countries where these are prevalent – in large part, this is exacerbated by regulation and accounting issues. They are inefficient because of
These regulations also introduce considerable uncertainty over future cash flow expense, which can only serve to lower current investment as companies prefer to hoard cash in difficult times. We would also note that DB pensions function as automatic stabilisers, while the alternate, DC arrangements, are intrinsically pro-cyclical.
Differences in the taxation treatment of various forms of pension should be eliminated. More traditional tools, such as, capital allowances and depreciation regimes, can be utilised for more targeted purposes in the corporate sector. One specific taxation variant that should be considered and costed would be the exemption from taxes of qualifying corporate infrastructure financing securities. We would see this occurring in conjunction with the elimination, more generally, of the deductibility of interest expense (see later). This could also perhaps extend to the issuance by corporates of qualifying securities where the income/coupons were exempt from income taxes in the hands of the investor.
Public Sector Infrastructure Agency
However, the most important range of policy options and instruments lies with the public sector. A specific infrastructure agency seems appropriate – largely to separate these activities from the issues of social policy and transfer financing. It might be hoped that this would break the association between government and low productivity growth. It is also clear that the predicted demands for investment far exceed the rather limited previous levels of government investment, and that capacity constraints arising from current limited government capabilities might well apply. This agency would be financed, not by government, but by capital markets. It should have the ability to issue tax-exempt securities, much in the manner of the US municipal market. Among the securities, we would expect it to issue, are perpetual income notes and deferred annuities. This would be the body, not sovereign states, extending guarantees and, in other ways, facilitating the external finance of infrastructure projects, both under government and its own ownership, and also in the private sector more generally.
Longer-dated Debt Issuance by Governments
One further policy shift might also be considered. Currently debt management offices issue the majority of government debt securities as “benchmark” issues in order to capture the liquidity premium as a cost advantage. This feeds into very active trading of government debt and a market culture that can only be considered short-term. It is, perhaps, superior for the debt management offices to issue targeted long-dated issues for long-term institutions, such as pension funds and insurers. Among these securities, we would envisage perpetuals and term annuities, including deferred annuities. It seems to us that there is a natural fit with deferred annuities for projects which have long construction phases before any usage revenues (real or imputed) become available.
On a more immediate note, it is evident that there are many current impediments which could be removed – individually minor, but collectively important. The fragmented nature of infrastructure commissioning among divisions of government and a lack of consistency in bid evaluation techniques and requirements is one. In this regard, the Laidlaw Inquiry (on the UK West Coast Line) conclusions are helpful as a guide to good practice: (a) ensuring future franchise competitions are delivered at a good pace based on sound planning, a clear timeline, rigorous management, and the right quality assurance; (b) creating a simpler and clearer structure and governance process for rail franchise competitions, including the appointment of a single director general with responsibility for all rail policy and franchising; and (c) ensuring we have the right mix of professional skills, in-house, and where necessary from professional external advisers.
The fragmented nature of the commissioning authorities carries the consequence that some of these have very weak credit quality. In elementary credit, it is often the case that the credit quality is considered of any entity is considered to be limited by the framework within which it is embedded. For example, many consider corporate credit ratings to be upper bounded , in principle, by the credit standing of the sovereign. While we do not subscribe to this simplistic analysis, we would note that the EU Project Bond Initiative proposes only credit enhancement of the project company financial structure. Both the letter of credit and subordinated debt approaches of the EIB operate within the capital structure of the project vehicle. While this may enhance the credit standing of that vehicle to levels acceptable to investors, it may not be the economically efficient route. In our opinion, the EU/EIB Project Bond Initiative should also consider offering credit enhancement to commissioning authorities. We note that many of the problematic private finance projects in the UK have had their difficulties rooted in the inability of the commissioning authority, the schools and hospitals, to continue to meet their debt service obligations while maintaining the quality and quantity of their primary functions.
There are also some residual cross-border impediments, such as the presence of withholding taxes in some cases – it appears that double-taxation treaties are being rigidly interpreted. To cite one example, a German Kapitalanlagegesellschaft (KAG), a common form of organisation of pension and other funds, is still subject to withholding tax on Belgian interest income. There are other withholding tax issues. It should be realised that the presence of withholding taxes and their administrative burden is a major drawback in an investment context, and particularly so for long-term investments.
It should be remembered that there are three areas of regulation that are relevant for potential policy action. The frameworks for infrastructure development, PPP or other, together with investment and financial markets, and the prudential regulation of the investment institutions are all relevant. In common with the OECD, we would see the task as being to create supportive scaffold across all three areas.
It may be useful to preface our response, with some published comments of Carolyn Ervin, a director in the Financial and Enterprise Affairs division of the OECD:
"So why don’t institutional investors live up to their long-term investing potential? Several complex and interlocking barriers hold them back....
"Institutional investors increasingly rely on passive investing or indexing on the one hand and alternative investments (such as hedge funds) on the other. The former can discourage them from being active share-owners while the latter may involve shorter term, higher turnover investment strategies.
"Agency problems are another barrier to long-term investment. Pension funds in particular rely increasingly on external asset managers and consultants for much of their investment activity. However, they often fail to direct and oversee external managers effectively - handing out mandates and monitoring performance over short time periods which introduces misaligned incentives into the investment chain. Institutional investors also contribute indirectly to short-termism via some common investment activities, such as securities lending or increasing investment in Exchange Trade Funds (ETFs). Investors may, therefore, be inadvertently contributing to speculative trading activities in the very securities that they own.
"Government regulation can also exacerbate the focus on short-term performance, especially when assets and liabilities are valued referencing market prices. For example, the use of market prices for calculating pension assets and liabilities (especially the application of spot discount rates) and the implementation of quantitative, risk-based funding requirements appear to have aggravated pro-cyclicality in pension fund investments during the 2008 financial crisis in some countries."
Different institutional investors have different goals, and these imply different horizons. Money-market funds may be institutional investors, but we should not expect them to purchase significant amounts of undated perpetual securities. At the other end of the spectrum, there are institutional investors whose horizons are indeed undated: university and college endowment funds and perpetual charities, for example, may take extremely long views in their investment portfolios. In the middle we may find pension funds and life companies, whose investment horizons will lie somewhere betwixt the two extremes, but may well have horizons of beyond 40 years.
The first way for these investors to play a greater role is to engage them in free discussion of their needs: it is only by recognising an institutional need for a particular type of investment that one can innovate productively. We would emphasise that this is engagement with institutions rather than their advisors and intermediaries. It should be recognised two of the principal long-term investor classes, pension funds and insurance companies are severely restricted by prudential regulation. Here we consider current accounting standards to be part of regulation. It is interesting to note that this prudential regulation is concerned principally with investor protection and, correctly, only to a minor extent with systemic stability. This is unbalanced; it does not consider the benefits, the positive externalities associated with the activities curtailed or prescribed.
Institutional investors, particularly pension funds, are currently poorly served by fund managers and investment advisors, whose advice is conflicted and often short-term in nature. In much the same way that a trader wants ever greater liquidity and trading, these firms have an interest in promoting the short-term, expedient and complex. Extending explicit fiduciary responsibility to them will greatly assist their institutional clients fulfil the long-term elements of their tasks.
The most important things that can be done for institutional investors is to remove the impediments which prevent these investors from pursuing their preferred investment habitats. These are overwhelmingly regulatory in nature.
The Green Paper considers a number of different forms of investment, for example, green and sustainable, infrastructure (among which are a multitude of types of investment) and SME finance. We do not find this helpful as the differences between these can be large and material for the manner in which they are intermediated. It seems to us that much of SME finance need is not long-term in nature. The predominance of competitive tender processes in the public-private model of infrastructure finance make this very different from the standard investment management model of insurance companies and pension funds. With standard investment, the pension fund or insurance company is concerned with the evaluation of an investment proposition presented to it. This is, in fact, how most green and sustainable investment propositions are marketed. The high costs and low probability of success of these infrastructure bidding processes are deeply problematic for insurance companies and pension funds. Indeed, it is possible to argue that this is trading, an activity that is explicitly penalised by the UK tax authorities for pension funds.
The risk profile of these investments is also unlike that of more traditional investments. With traditional investment, risk grows with time; the concerns expressed by institutional investors over consistency of political commitment and regulatory stability are similar in nature. However, the concerns over procurement and construction costs for infrastructure, the high failure rate of new and small companies (fifty percent of newly incorporated firms do not survive their fifth birthday in the UK), and uncertainties over the viability of green and sustainable investments at scale in the absence of subsidies add a new dimension. Here the risk is loaded onto the short-term rather than growing with time.
The question of developing a larger market for institutional private placement for debt securities is under investigation by the Association of Corporate Treasurers at the request of the UK Business Bank. This market would hold the prospect of lower volatility of prices as valuation could be based upon economic fundamentals rather than market prices.
The problem currently lies with the present form of prudential regulation that has simply been erroneous in writing across from banking a form of risk-based regulation that does not recognise the different nature of long-term institutions (and that failed even for banks when tested). One of the distinctions that should be drawn is in the role of maturity transformation.
For an institution, such as a bank, maturity transformation as a viable business strategy is dependent upon future funding availability. If the loan is negotiable, it may alternately be sold. If the deposit advanced in a loan is subsequently called, it must be replaced. In the case of markets, there is a similar maturity transformation role - the equity or bond, bought in the expectation of exit and realisation by sale in the market at the future then-prevailing price, are obvious examples. In fact, if an asset cannot be priced in a market, it is usually impossible to use it as collateral in, for example, repo financing. In this regard, we would draw attention to the fact that the August 9 2007 ECB intervention in markets, which most regard as the seminal event of the crisis, was triggered by the inability of a bank to price securities in several collective investment funds. In this regard, bank funding (liquidity risk) and market liquidity are closely related concepts. This is a material risk and correctly a prudential concern.
Insurers and long-term assets
Stylistically we may describe banks as having liquid liabilities and illiquid assets, while insurance companies have illiquid liabilities and liquid assets. However, there really is no reason for any requirement for long-term insurance institutions to hold liquid assets. This may be prudent in the case of general insurers where substantial claims (or liabilities) may become payable at short notice, but for most classes of long-term business this is not true.
For an institution that has liabilities that are long-term in nature, there need be no such maturity transformation risk. For example, an insurance company may hold (even unlisted) equity among its assets up to the amount of its own equity and retained earnings without having any maturity transformation exposure. It is also important to understand that many of these long-term institutions are not exposed to market risks – their liabilities cannot be subject to a ‘run’ in the manner of bank depositors. Surrender terms, where there are any, are typically punitive. In the long-term, prices should be immaterial to these investors – income and fundamentals dominate the security of their liability contracts. The relevant consideration is the gap between asset cash flow maturities and the equivalent liabilities.
Prudential regulation is risk-based; the value and accuracy of these risk estimates have been widely questioned. The regulatory imposed risk weights have also been widely, and validly criticised. The point which has not received as much attention is that the remedy under these risk based systems is unique – extra capital buffers or provisions, currently held. We would point out that one of the few things we know (with certainty) about risk is that it means that more things may occur in the future, than will. The consequence of this is that it is only too easy to over-provide, which will limit the efficiency of the institution and the savings they represent. We would point out that these buffers may be needed only in the future, which suggests that provision today may be time-inefficient. In fact, the standard solution to this problem in other situations is insurance – we routinely insure against fire, flood and other risks. We would stress the point that it is perfectly possible for a financial institution to self- insure against future risks through the structure and form of its asset portfolio. Such institutional insurance arises from future non-market cash-flow, but is rarely if ever recognised in the market prices of the institutions. We would emphasise that the current and proposed regulation of banks, insurers and pension schemes is a monoculture – and a monoculture that is firmly rooted in the here and now of market prices and current buffers. This is a source of systemic instability.
This is greatly compounded by current accounting standards (see later). It is nonsensical that current regulatory standards focus upon the present position with respect to the institution’s solvency. For long-term institutions, the composition of the assets held is all-important with respect to the ability of these institutions to discharge their obligations in full and on time in the future when they fall due. Under the current standards, an institution may hold exclusively cash again the present value of liabilities and be considered sound. However, it is clear that, without earnings that are at least equal to the discount rate applied to actual liabilities, the institution will be unable to discharge these fully and on time as they fall due. The approach in regulatory use may be described as balance sheet insolvency when the reality in markets is equitable insolvency. This latter form requires the existence of an uncured payment default, before acceleration occurs. To use discount rates derived from market prices or yields is inappropriate, but it is part of the current regulatory valuation process.
The accounting standards are also mixed attribute in nature. Assets are valued at market prices, while liabilities are discounted present values. These introduce material bias and volatility into pension and insurance accounts which is unreal – the bias and volatility are an artefact of the accounting measurement system rather than of the institutions being measured. Above all, when setting the level of prudence required, regulation should recognise the benefits that will be foregone. For example, insolvency for DB pension schemes has been a rare event, but their regulation in pursuit of pensioner protection has effectively closed this form of pension provision, which has served millions of pensioners very well over many decades. The substitutes now offered – DC arrangements – are massively inferior and in a number of respects may be economically harmful. There are further questions that need to be considered when inefficient forms of saving are adopted, particularly when these savings are tax advantaged.
We are also concerned that regulation across different types of institution can be inconsistent. In this regard, we would point out that, for banks, the absence of long-term liabilities is penalised, while for insurers it is their presence.
In most product categories, residents of the various member states are faced with markedly different marketplaces in terms of the product offering, prices, costs as well as the availability and nature of information. There is clearly a tax and regulatory-induced home bias. There has been some consolidation among market intermediaries, but that has raised more competition issues than it has reduced fragmentation across states.
Investors are not able to access similar products in all member states. Some product providers, operating in an inherently national regulatory framework, are prevented from offering similar products in other member states. There are concerns over a lack of comparable information and insufficient transparency about fees and costs, as well as limits to the transferability of savings/investments from one product/provider to another (exit fees, legal barriers).
The question of transferability is much misunderstood. For example, the question of pensions transfers where the question of transferability is known as portability. The primary concern in many cases in not portability but in fact preservation. There really is little or no need for a pension literally to follow the member, what is required is preservation of the accrued benefits.
Transferability is hardly problematic when the pension has individual DC form, as this is, prior to retirement, little more than a tax advantaged savings scheme. For DB schemes by contrast, there may be material differences between schemes and any transfer process more complex to achieve equitably. We feel that the detail of these issues and particularly the question of a common platform should be the subject of a separate consultation, as they are not confined to long-term investment.
We have already indicated the use of tax-exempt securities or contracts, but there are many other methods. These all involve forms of risk sharing or risk-pooling among individuals and with the institution. Mutual organisation has much to recommend it in this regard.
It is important that long-term contracts participate in the performance of the institution or the economy. The ‘with-profits’ policy of pension and life insurance was a classic contract design. Similarly, the traditional UK defined benefit pension scheme was a very efficient form of contract design – highly suited to purpose from the perspective of the beneficiary.
These institutions enhance the financial depth of an economy, which here means that fewer savings are required, to achieve a specific objective, than might be the case in their absence. They also reduce the need for an ongoing individual involvement, though the individual is fully committed.
We would suggest that market consistent accounting and regulation, which rely upon implicit or explicit replication, would be deficient in this regard. Bundles of market-traded securities cannot be expected to reflect the synergies of risk pooling and risk sharing which are contractually possible within and with such institutions. Put another way, these forms of institution exist because their contracts cannot be replicated in other ways.
We would expect the proposed ring-fenced and utility banks that will be dealing with retail depositors to operate as both advisor and broker of mutual funds and direct investments to their wealthier depositor clients.
It is almost impossible to exaggerate the cumulative effects of current regulation, which includes, for these purposes, current accounting standards for long-term liabilities and investment. Proposed regulation might, not unfairly, be described as more of the same. It has had dramatic effects on the degree of pro-cyclical activity. Not only should we be concerned with explicit institutional fire sales, there is a more important question of the conditioning of the markets.
EDHEC Business School studied the effects of the proposed Solvency II regulations on bond fund management for insurance companies extensively. As can be seen from the illustration below , there are significant capital costs associated with bonds that increase markedly with declining credit quality and with increasing term. The magnitude of these risk weights will serve both to limit the amount of debt finance available and provide an underpinning to its cost.
The asset allocations of long-term insurance companies and of many pension funds have moved markedly towards more bonds and other instruments that lie well below their optimal or design risk tolerance and risk-bearing capacity. These investment policies are acutely limiting the terms of many consumer contracts, such as annuities.
Derivatives instruments, which, it should be remembered, are contracts inside of the financial services industry, have found wide usage in response to regulatory pressures. They can be expected to have little effect upon outside activity, including investment, beyond the rents charged, so effectively, upon them by the financial services sector.
Even at the level of institutional design, procyclicality is induced. The shift to individual DC pensions is an illustration. Traditional DB is contra-cyclical; it provides automatic stabiliser effects to the economy. DC, by contrast, is actively pro-cyclical. Behaviour induced by individual wealth effects matter in the case of individual DC – including the lowering of interest rates in response to declines in economic activity. Though many regulators have stated they are exercising forbearance with respect to the funding status of pension funds, the fact is that many schemes have been required to lower benefits, for example, in the Netherlands, or contribute much more (UK). In the UK, some life insurance companies have been required to lower equity asset allocations to satisfy their regulator.
The Anomaly of Liability Mark-to-Market
One present accounting nonsense is that the liabilities of an institution decline in line with the market price of their obligations. The classic illustration of this was the “profits” taken by major investment banks during and after the crisis, when their continuing existence and service of their debt obligations was in doubt. Unless the body has bought back these obligations at their discounted price, the obligation to service and repay those obligations is unchanged by their market price. The obligation is a function of the terms on which the liability was incurred by the institution. The market price reflects also the likelihood that the company may not make those payments as well as the state of liquidity in the market in which they are traded. The concern for the institution is with meeting the terms and conditions under which the obligation was created. This is a question of its own current and future liquidity. Perhaps it is easier to see this in another context. The payments an individual must make under the terms of their mortgage do not decline (or increase) with the market price of mortgages securities.
The mixed attribute nature of pension accounting is also a concern – the use of market based discounted present values for liabilities and market prices for assets leads to biased and incorrect results. This was investigated in depth in a 2013 paper by Keating, Settergren and Slater entitled Keep your lid on!. That paper illustrates the way in which a form of amortised cost accounting may be applied to DB pension schemes in a manner that is fair value consistent. The paper also demonstrates that market consistency is not fair value consistent. It is notable that the International Association of Insurance Supervisors’ 2011 Insurance Core Principles, Standards, Guidance and Assessment Methodology states in Section 14.0.4 which deals with valuation that: “An economic basis may include amortised cost valuations and market-consistent valuations”.
We would also contend that market prices constitute a very poor basis for decision and that the concept of market discipline, a feature of most regulatory systems, is badly flawed in respect. The critique is that the signal in market prices, which is relevant to long-term fundamentals, cannot be distinguished from the noise. We have illustrated this point (using a simplifying assumption of normality of returns) for ranges of return and noise on an annual basis, showing the number of years at which the signal will be equal to the noise – only in the light blue area is the signal greater than the noise. Moreover, when market prices are dominated by short-term participants, we should expect that market discipline will reflect their preferences rather than the preferences of long-term investors or holders.
This question of incorrect signals from market actions is directly observable. The RBS take-over, which proved so disastrous subsequently, was explicitly and resoundingly approved by voting shareholders.
We are repeatedly cautioned that we should not read too much into particular outcomes in most other aspects of our daily lives, but here, we have precisely the opposite being required by accounting standards and much regulation. It is particularly pernicious in that the academic hypothesis, which attributes so much to market prices, has the universal ‘get-out’ clause that these prices reflect or discount some risk or other, which of course is unobservable. It is not a testable, and possibly falsifiable, financial theory. The related question of forecast errors is accessibly, and well covered in an eponymously named speech by Ben Broadbent, of the Bank of England’s MPC.
Currently equity capital markets do not constitute a meaningful source of new finance for private sector investment; they are more important as mechanisms by which existing entrepreneurs may realise their interests. They now serve (poorly) as (costly) governance mechanisms. By contrast, the debt capital markets are important sources of finance for industry and government. Recently, this has been extended to smaller companies by exchange based listing and trading of debt securities. Retail investors are reportedly active in these markets.
The Role of Credit Rating Agencies
We note the role of credit rating agencies (CRA) in debt security evaluation and their inclusion in the regulatory regimes of financial institutions. We feel that these requirements – for securities to be rated – should be eliminated. These requirements give the ratings agencies a role in determining access to markets, without concomitant responsibilities and duties. Rating should be an entirely voluntary exercise on the part of an issuer – a decision to be taken in the context of costs of issuance with and without such ratings. With adequate disclosure standards ratings agencies are entirely redundant. It is also a mistake for such ratings to be used in the risk evaluation of investment funds. Some have noted that this is a problem of collective organisation and have suggested that individual provision, with products such as DC ‘pensions’, may be superior. All that is achieves is an attitude of ‘après moi, le deluge’ and extremely volatile unstable financial markets. The greater issue is not the defects of particular forms of collective organisation, but rather that of the resulting co-ordination failure.
The Generational Problem in Pension Funds
The greatest issue in equity markets is one of governance. Many reports, such as Kay and Cox, have highlighted this. Here, we would not want to leave the impression that we believe that governance is the most important driver of corporate behaviour; that, of course, is competition.
The governance question, in essence, is one of the tragedy of the commons, where distinct generations have different incentives with respect to usage. This can be illustrated in the context of pension funds. Here the members close to or in retirement have a greater interest in preserving capital values rather than risking their entitlements alongside younger members – particularly when a scheme is underfunded. Wrongly, many have taken this to be an argument for extremely conservative management of pension funds. As with a two generation commons problem, the solution is to offer differential control rights. In the case of the pension fund, giving control (or weight) to the younger generation will align the generational interests. The beneficiaries who have the claims the most remote in time should have the greatest proportional say. The younger generation will wish to accept sufficient risk to achieve their objective of receiving the promised pension, but will not take excessive risk as this would simply result not only in harm to current pensioners but destroy their own property. This is a balancing act to be performed by scheme trustees. In the case of corporate occupational schemes, there is also the position with respect to the sponsor to be considered.
There is a related issue of trust versus contract. The superiority of systems based upon trust can be demonstrated at the macro-level by the superior returns (usually) achieved by trust rather than contract based pension schemes.
Rewarding Long-Term Equity Investors
Differential control rights, or voting entitlements that increase in the length of ownership could go far in aligning shareholder interest in the long-term. This is not simple to achieve in voting rights for traded securities, but is feasible from an incentive standpoint in dividends and other distributions.
The Trend Away From Equity Markets by Long-term Investors
The current state of markets is poorly understood. The average savings institutional holding period has not declined as dramatically as might be construed from overall market turnover, which has increased manifold. What is happening here is that the trading activity in the ‘free-float’ of equity securities has multiplied incredibly. Much of this has been the result of the advent of high frequency trading. It should be realised that trading or the need for the option to trade is a sign of a lack of commitment.
The UK serves, perhaps, as an illustration of recent trends. According to the Office for National Statistics, UK equity holdings of insurance companies have fallen from 20.8% of the London market in 1991 to 8.6% in 2010 and for pension funds the situation is even more pronounced – from 31.3% down to 5.1%. Unless these long-term holders have been replaced by other long-term holders, the London stock market has become a venue for far more maturity transformation than was previously the case.
The Erosion of Trust and the Rise of Algorithmic Trading
There is a related issue – that long-term holders sell some or all of their holdings to short-term speculators and “arbitrageurs” when take-overs or bids emerge. These short-term actors usually have an interest in ensuring that the bid or other transaction ultimately occurs. We see this behaviour on the part of long-term institutions as being motivated by a desire or need to participate in short-term performance – for example, under third party mandates awarded to them. Selling by long-term institutions should not be considered as unwise or even anti-social, given the radical nature of these changes and indeed, the rather dismal track record of successful mergers and acquisitions.
It is clear that trust in market fairness has broken down – many institutional investors now utilise algorithms to execute their market activity precisely because they have lost trust in their ability to do this without exploitation. This market impact facet of institutional activity is a significant contributor to the effective costs of portfolio management . There is also a further effect. Markets that are dominated by short-term noise traders, of which computer-based algorithmic traders are a form, will be less efficient than markets where this is not the case. The informed investor will dominate activity in the less liquid stocks. The highly liquid actively-traded stocks will tend to converge more slowly than these illiquid stocks to fundamentals.
It is true that there is a role for market-making intermediaries to overcome the problems of outside order imbalance – usually referred to as liquidity provision. However, it is also true that too much liquidity may be available. It is notable that the liquidity provision role offered as defence by the high frequency traders does not address why it is that they do not dedicate their capital to liquidity in the low liquidity stocks, where the premium is highest but rather focus upon stocks those that are already liquid. Equally it does not address why liquidity dries up instantaneously whenever there is any sign of market stress or distress. There are a number of practices associated with high frequency trading which we would like to see suppressed. For example, the practice of entering literally thousands of orders with the intent of withdrawing these unexecuted gives a false impression of the depth of a market. A financial non-transactions tax would reduce or eliminate these.
Capital markets will have difficulty filling the equity gap. As long as regulation favours debt over equity in portfolio allocations, this will not occur. Similarly, as long as the tax treatment of interest cost favours debt issuance, corporations will offer debt securities.
In fact, the equity gap is most pronounced in early stage investment and this is where the information asymmetries are largest. Indeed, there is much that is speculative about the start-up and early development phases of most businesses. It is not at all obvious that these start-up companies have any natural role in the construction or operation of infrastructure although they may do so in ‘green’ and sustainable fields – in, for example, spin-offs from university research. However, the information asymmetry issues will severely restrict the utility of incurring the additional costs of seeking public listing. These ventures are best handled in private bilateral negotiations. In many jurisdictions, there are already tax incentives in place to facilitate these. It is possible that further pooled vehicles could be created which would mitigate many of the information asymmetry problems, and indeed that many such as the UK EIS and Venture Capital Trusts (the European VCT appears a good start here) already exist.
We believe that one of the problems of regulation is that the voice of intermediaries and their trade associations has been loudest and most strident, and has overwhelmed the voice of long-term investors. The intermediary community have a natural interest in exchange, not investment: this is intrinsically short-term.
This survey of the covered bond market within the context of the question focuses greatly on specific recommendations. In the interest of space we have been parsimonious in our reply, but can expand at length on any specific question if asked.
European covered bond legislation has been the global standard setter for this asset class. In order to revive this position after European covered bonds (originally a AA+/AAA asset class) lost some global market confidence, some harmonization and improvements are necessary. To ensure affordable mortgages, broad cross border investor acceptance of covered bonds is necessary not only within the EU, but globally. Good covered bond legislation and regulation should be easy to evaluate by both domestic and foreign investors.
The mortgage bonds of Fanny Mae and Freddy Mac bonds are not only the global competitors to European covered bonds. The concept of a centralised refinancing hub for mortgages with an implicit government guarantee similar to FannyMae and FreddyMac has been proposed. From a market development and regulatory point of view, the alternative is the decentralized, high quality legislation with many competing market participants. Part of the question is therefore, how much harmonization is required to compete successfully against the centralized US model in the global fixed income market?
Some diversity of covered bond legislation across Europe offers investors a choice. Full harmonization could lead to a US-style centralized model, which would probably not be able to compete well with the US institutions and their track record of full government support. However, too much diversity makes the analysis of the assets too complicated and rather costly. The core areas of harmonization that should be considered are:
Clearly, securitisation has a place in European capital market in the future. The constrained lending capacity of the banking system alone is sufficient reason. However, the experience of the crisis with this type of investment is fresh in the memories of many investment institutions, and it will take a long time to overcome this.
The ‘originate to distribute’ model was undoubtedly viable, though subject to moral hazard. This can be overcome by retention of a significant first loss exposure by the originating institution. This is the equivalent to the ‘deductible’ of standard insurance policy pricing.
There is a second issue, which concerns the level of maturity transformation inherent in the financial structure of the securitisation. This often depends upon the presence of significant levels of financial engineering with respect to liquidity provision within the service structure and with respect to maturities. From a systemic standpoint, these securitisations may constitute a material level of risk, and of course, these engineering derivatives contracts create contagion between financial institutions and markets.
It is not at all obvious that the presence of insurance mitigates these issues, though where this takes the form of subrogation rather than financial guarantee, it is clearly more feasible. Note that the original insured municipal securities in the US market were insured as subrogation contracts, where the insurer stepped in the event of a default and assumed responsibility for the payment of coupons and principal under the original terms of the security. Financial guarantee, with its lump sum payments in settlement at default, were a later development. From the perspective of the long-term investor, the subrogated form is superior as it continues the performance of the investment under the terms and conditions under which it was acquired, thus causing the least disruption to the investment structure. This shift away from subrogation is another incarnation of the short-term versus long-term problem.
One way in which the securitisation market could be initially revived would be for the development banks to create and offer these vehicles. These development bank instruments could also be explicitly treated as eligible instruments for central bank open market operations. The precise terms of issuance – retentions, overcollateralization and liquidity provision, as well as the financial structure of the vehicle are a question of detail at the time of issuance.
With a number of such securities outstanding and placed with investors, it would not take the private sector long to begin issuance, having been guided by development bank practice. We do not see a significant role for rating agencies in this process, nor would we like to see credit ratings embedded in guidelines or regulation.
The specific savings account that we would like to see more prominently issued is the nonnegotiable savings bond available to retail investors. This instrument is, in fact, the “risk-free” of financial theory. In theory, it should yield more than marketable government securities and not, as has often happened in the past, be an instrument of financial repression.
As was once the case in the US, it should be possible to collect these investments alongside wage payments under specific contractual arrangements. In addition, coupon/interest payments should be immediately reinvestable into new savings bonds.
These instruments could be long-term in nature and include both immediate term and deferred term annuities. We would not propose that this issuing entity should assume longevity risk. With the increasing emphasis on DC pension provision, there will be considerable growth in demand for annuities of all types.
We have not considered the relative merits of the many savings accounts in the European member states. We do not see merit in an EU-wide model. These savings accounts reflect national preferences and their tax structures. An EU model could be expected to have different costs across member states.
The deductibility of interest payments from income is one of those situations that would never be created if we were starting from a tabula rasa. It biases corporate financial structure towards debt and with that, contributes meaningfully to the cyclical nature of an economy. This question was considered by the IMF. In his 2011 paper, “Tax Biases to Debt Finance: Assessing the Problem, Finding Solutions”, Ruud de Mooij states:
"Most tax systems today contain a “debt bias,” offering a tax advantage for corporations to finance their investments by debt. This has grown increasingly hard to justify. One cannot compellingly argue for giving tax preferences to debt based on legal, administrative, or economic considerations. The evidence shows, rather, that debt bias creates significant inequities, complexities, and economic distortions. For instance, it has led to inefficiently high debt-to-equity ratios in corporations. It discriminates against innovative growth firms, impeding stronger economic growth. Debt bias also threatens public revenues, because it enables companies to reduce tax liabilities by using hybrid financial instruments as well as by restructuring their finances internally, moving debt between affiliates.
"These traditional distortions of debt bias have long been recognized. Yet, recent developments suggest that its costs to public welfare are larger—possibly much larger—than previously thought. Companies appear to be responding to the incentives of debt bias more over time, so the associated cost to public welfare has been rising. The economic crisis has also made clear the harmful economic effects of excessive levels of debt in the banking sector, especially due to the systemic effects of bank failure. These insights make it more urgent to tackle debt bias by means of tax policy reform."
We note also that the recent G30 publication, Long-Term Finance and Economic Growth, makes an explicit proposal with respect to removal of CIT relief:
“Proposal 11: Remove the bias against equity in countries where it is present. Long-term finance can be provided in either debt or equity form; both methods have an important role to play in a resilient system of long-term finance. At present, however, many tax systems create a bias against equity investment. The case, in principle, for removing this bias is widely accepted, but little action has followed, reflecting in part the inherent difficulty of changing tax regimes without creating windfall winners and losers. However, the arguments in favor of action are sufficiently compelling that governments should seriously consider the options for national policy initiatives, and the case for coordinated action should be on the agenda for international discussion.
"There are a number of policies that could be implemented. The double taxation of equity should be removed in countries where it exists. In advanced economies with mature debt markets, policy makers should consider changing the tax treatment of debt and/or equity in order to remove the bias, while also ensuring that policy changes are revenue neutral. One option would be to eliminate the tax deductibility of interest payments at the same time as lowering the marginal tax rate for corporates. An alternative option would be to apply tax deductibility to equity dividends at the same time as increasing the marginal tax rate. These options are presented as suggestions that require further discussion.”
We do not make any comment on the merits of these proposals but would note the deductibility of interest expense does have the advantage that it makes the operation of management of economic activity by monetary policy easier in some circumstances. It increases markedly the dependence of commercial and industrial activity on bank financing – a more diversified capital base would be superior. In other times, the replacement of the assets for the banking system might be problematic, but at a time when the banks are reducing their overall balance sheet size, this may be a positive. As it would operate selectively, with strong companies raising more equity and capital market finance, it would leave the banks with a greater emphasis on smaller companies.
As with so many things, the problems come with the transition to a world where no further deduction was available. However, that said, it would make the effects of any tax concessions associated with infrastructure and sustainable investment all the more powerful. If the interest on qualifying sustainable and infrastructure investment debt financing were deductible and this was combined with the municipal type treatment of securities offered to investors, the cost of funds would be markedly lower than for equity or debt under its current treatment.
Doubtless, there will be critics of this idea who will cite Modligiani-‐Miller and assert that lower returns to equity for the corporate sector will result from the cessation of tax deductibility of interest payments. It has long been known that there are empirical problems with the Modgliani-‐Miller theorems, as has been recognised by Merton Miller In fact, the empirical evidence is that, contrary to the predictions of this theorem, unlevered firms persistently achieve higher profits than levered firms. The margin commonly cited is substantial – of the order of 5% in the US. Note these are the reported profits of companies, not market returns.
In pursuit of our brevity objective, for the question of which form of CIT reform is to be preferred, we refer the reader to the earlier cited IMF paper and, also, to chapters 17, 18 & 19 of “Tax by Design: The Mirrlees Review”. We feel that this matter should be the subject of separate consultations.
This question cannot be answered in a general way. The particular considerations depend upon the particular circumstances of the particular country. These include the mix of sources of government revenues, its social welfare system and infrastructure investment needs, as well as the structure and degree of development of its financial system and private sector production.
We would point out that across Europe, not only do the levels of household wealth vary markedly, but so also does its composition. In part, this is historic, cultural and a functional of the national endowment, but in part, it is, also, a result of national tax and social security systems. We illustrate this below for Holland and France. We would first note that land and dwellings account for 80% of wealth in France but only 43% in Holland. In addition, the composition of financial wealth is markedly different:
There is one issue that, we believe, spans the European community. This is the question of intergenerational equity. Often expressed as “we are stealing from our children” and usually accompanied by cautions that this will result in intergenerational strife and perpetuation of a growing level of inequality. This is an important aspect of long-term saving and sustainability.
We do not accept some of the more outrageous assertions, such as those concerned with the level of debt to be inherited by our successors. We may be leaving an average debt of, say, £25,000 to our successors, which they will have to service and repay, but the question unasked by these doomsayers is to whom they will be repaying that debt. We do accept that there may be issues concerned with sustainability.
The most obvious solution to these problems lies in a wealth tax. Since 80% or more of our wealth is owned by the over-50s, this would is age-selective taxation. Moreover, it could also be sustainability friendly, if sustainable investment wealth was exempt for the purposes of the tax. For greater detail on wealth taxes, we refer the reader to “An Annual Wealth Tax”.
Finally, we would add that there are interactions with other parts of the social security and welfare system, for example, the means testing of benefits. The tax exemption we have proposed would be from taxes on income (and perhaps capital gains), but not necessarily from the income and wealth assessments of means tests.
We will not address this question in its complete generality, and just content ourselves with the observation that a beneficial tax incentive is one that results in activity that would not have otherwise occurred, or would have occurred in a manner that was more costly to the social welfare. From this, it also follows that the tax cost should be less than the cost of remediation if a project is undertaken in non-sustainable fashion.
We have indicated that we would like to see the debt over equity bias in taxation eliminated and that debt raised for sustainable and infrastructure investment by a corporation might be exempted from this. The question of arbitrage then becomes one of mis-classification.
Obviously, the company’s auditors have primary responsibility for ensuring that the tax treatment is compliant with legislation. Notwithstanding this, the classification is subject to agreement by the revenue authorities. We would accompany this with the ability of the revenue to levy fines and other charges on both the corporate and its auditors in the event of misclassification. In cases of doubt or ambiguity, for example, where only part of an investment programme was sustainable or infrastructural in nature, we would envisage that the corporate would be encouraged to consult the tax authorities. Of course, this would also apply to both sustainable investment exempt under the wealth tax regime that we believe should be considered as well as to any securities issued under a municipal-type income tax exemption for investors.
No, not if the form is regulation. The EU could provide a forum for discussing and co-ordinating activities, but the format would have to be co-operative among the member nations. Regulation would achieve little or nothing. It is in fact out of sequence. Before considering taxation, it would be necessary, as a minimum, for co-ordination of fiscal policies to have occurred. It evident also that there are many issues still outstanding with respect to prior EU policies; these compliance issues should be resolved before considering any further steps. A matter of building on a solid base.
It should be remembered that the composition of government revenues varies markedly across nations, as do their expenditure preferences. Moreover, the yields from and the efficiency of particular forms of tax and duty vary markedly across member states. An imposed variant would run the risk of imposing unwarranted and differential costs on member states.
We wonder if there may be a place here for the behavioural sciences to be brought to bear. The UK Cabinet Office’s behavioural insights team (“Nudge Unit”) has reported some remarkable successes in achieving tax and regulatory compliance, albeit from individuals.
In summary, we would see tax co-ordination at this time as being neither feasible, nor desirable.
Fair-value accounting involving a regular mark-to-market has had a nefarious effect on investor behaviour. We do not believe that compensation or correction of its negative effects is appropriate. The problems are most pronounced in the accounting of long-term institutions. The liability issue is well recognised. We would like to see long-term financial institutions exempted from the market consistency regime. Returning to the distinction drawn above between speculation and long-term (i.e. patient, productive and committed) investment, mark-to-market accounting may be appropriate for securities trading, but it clearly is less so for long-term insurers or pension funds. We note that the illiquid loans of banks may be classified as “held-to maturity” and valued on an amortised cost basis.
For assets, the question at heart is the function of the securities held. For a dealer they are a finished product, available for sale. For an insurance company or pension fund, they are an input towards the finished product, the future cash flows that are the pension. The selection and management of these financial products is the production process of the insurance company or funded pension scheme with this end in mind. Sale of some or all of these securities held may be contemplated as part of this management process, which operates over time. Mark-to-market, when combined with risk-based solvency regulation, can disrupt this. It can be expected to result in strategies that satisfy the regulations in force, rather than optimise the management production process. This is costly to the consumer – it is short-term.
We have already criticised the idea that market consistency is replication valuation. If an institution’s operations could be replicated by bundles of traded securities, the institution would be entirely redundant. Market consistency is often promoted as being fair value consistent. The fact is that it isn’t. This point is demonstrated in the paper “Keep Your Lid On”.
The issue is compounded by insistence upon the use of liquid traded securities prices, which themselves contain liquidity premia. Liquidity has a cost, if it did not all assets would be liquid. Thus liquid securities are the securities which are least informative with respect to fundamental value and to a very large extent irrelevant to the value of these securities in their use.
The use of discount rates derived from bond market rates to derive present values is entirely without theoretical support. Moreover, no insolvency court could be expected to admit as creditor claims these values. These values are amortised costs, or equivalently the discounted value of the expected payment(s). This distinction is important. Consider two zero coupon bonds with the same maturity, five years from the valuation/insolvency date, issued by a company at different points in time, one with a yield of 10% and the other 5%. The admitted claim amount of the 10% issue would be 62.1% while the 5% issue would be 78.4%. Moreover, markets recognise this difference; post insolvency, these bonds would trade at prices which reflected these different claims amounts – in the ratio of 1:1.26.
There is much confusion over the question of insolvency and acceleration. The admitted claim on a long-term liability will not be the full ultimate value of that liability but rather the amortised cost of that liability. With coupon paying bonds that is the original principal plus accrued coupon to date. Many of the problems of the pensions world have arisen from the desire to protect some classes on pensioner, for example, by meeting all of their claims in full, which penalises active members. It should be realised that the active member’s claim is the one that is most remote and may have been issued on entirely different terms from the long-retired pensioner member.
The accounting practice of comparing discounted present values for liabilities with market prices for assets (“mixed attribute” accounting) breaches elementary measurement theory on several grounds. Theory would require us to use the same measure for both objects to be measured and it would require that these measures should also be invariant over time.
Our proposals for pension scheme valuation and accounting are contained in “Keep Your Lid On”, and accompany this response. It is trivial to extend these techniques to long-term insurance policies and indeed this would be entirely consistent with the amortised cost method recognised in the IAIS Insurance Core Principles.
Our earlier reference to solutions to the tragedy of the commons indicates the manner in which incentives may be designed to promote shareholder engagement (note that we prefer the expression commitment). Long-term shareholder engagement can be promoted through two main channels. These are: (i) Control Rights and (ii) Dividend Policy. The key insight from the commons representation is that those members whose claims are most remote in time should, ceteris paribus, dominate the decision process. To illustrate this, the youngest member of a pension scheme should have the pension rights that are most remote in time. This is not those who have held their claims the longest, but those whose claims have the longest remaining time to performance.
It should be emphasised that shareholder commitment is not altruistic behaviour. It permits management the time and resources to invest in other stakeholder relations, which add value to the enterprise, and this value accrues to the benefit of shareholders.
That said, we believe that no further steps (not already envisaged in the Action Plan) in this direction should be taken now: too many initiatives underway simultaneously are extremely difficult to coordinate and could easily result in havoc. We should execute the Action Plan on European company law and corporate governance, and examine carefully the results of previous and current initiatives.
In recent periods several initiatives have been implemented or are currently being discussed that are intended to promote long-term shareholder engagement. The most evident regulation is the Shareholders’ Rights Directive (2007/36/EC) that was implemented only recently. Time is needed to observe informative, complete results, and this is a long-term process. We should not expect immediate results, but there are already some early signs of increasing shareholder activity.
There are some pitfalls in some of the proposals circulating. Many of these problems stem from the distinction between shareholders who have owned for a long time, now past, and those that will own shares for a long time, in the future. The future problem is essentially one of time consistency. This is a question of the investor’s ability to commit credibly to investment for a future period. Control and dividend rights based upon past holding periods may reward the historically committed shareholder, but they may do little, or nothing with respect to the future issue. Indeed, they also would not have materially increased the ability of management to add value by investment in and development of relations with other stakeholders in the past.
The proposal of paying special dividends to long-term shareholders is seen to be problematic by some. These look to the OECD Principles of Corporate Governance (which applies both to increased voting rights and to increased dividends), where Principle III.A requires that “All shareholders of the same series of a class should be treated equally”, and interpret this as ‘any special privilege within the same series of a class would go against basic corporate governance principles’. If this interpretation is correct, then it means that special classes of shares would be created by these actions and with that many other difficulties. However, there are a number of listed companies where holding-period related dividends and bonuses are paid. One example is Air Liquide, whose shareholders’ charter, inter alia, states:
A company that has consideration and respect for all its shareholders:
A company that remunerates and increases investment value over the long term for its shareholders.
The interpretation here is that all shareholders have equal rights – a form of equality of opportunity. It is the action of the shareholder which maximises the value of the particular shareholder’s interest. There are no problems with sale in a market - though the seller would be foregoing any benefits accrued over the period held. Benefits that will not be reflected in the share price – though the share price will reflect the presence of this incentive structure. No multiple classes/series of shares have been created, nor additional records required. The shareholder’s register will have recorded the date of entry. The embedded value associated with such accrued rights will tend to make sale in a market less attractive than might be the case with standard arrangements. This is a form of effective barrier to exit, and will tend to increase the commitment of shareholders to the company.
We do not believe that proposals for granting increased voting rights to long-term shareholders are well founded. There are several dangers connected with such a tool. The most important factor is uncertainty over the total number of votes assigned at a specific time. Shareholders or other investors including market participants would not know the aggregate voting population or the degree of influence of their particular voting entitlement. The result would be problematic in respect of some obligations specified in the Transparency Directive and Market Abuse Directive. Such proposals do indeed seem a prima facie breach of the OECD principles evoked above.
By contrast, in mutual organisations, it is possible to offer differential voting rights. The young with-profits policyholder should have a greater say per unit of wealth committed than the maturing policyholder should. This extends, in general, to member mutual arrangements, such as many industry-wide Dutch pension schemes, and the UK civil service scheme (albeit that is unfunded). In general, with access to pensions savings restricted prior to normal retirement age, commitment is more credible, and the problems associated with market sale do not exist.
There is one current practice that we would like to discourage or suppress: the return of excess corporate cash to shareholders by share repurchases rather than dividends. The differences between dividends and stock repurchases are material in a number of ways. All of the usual accounting-based performance metrics are unchanged, with the exception of earnings per share (EPS). In the case of a buy-back, EPS increases, while a special dividend reduces EPS. Therefore, it really is not surprising, given the prevalence of management option incentives, for management to favour buy-back over dividend. Whether share buy-backs add or subject value for remaining shareholders depends upon the price of the shares repurchased. Those bought below intrinsic value will add value, while those purchased at premiums will subtract it. This view is supported by a number of empirical studies which have considered whether buy-backs may be justified by the subsequent valuation and market performance of the remaining shares.
Attitudes to buy-backs have changed. Forty years ago, they were taken as an indicator of a lack of management confidence in the prospects for a firm’s business. Now they are taken to be expressions of management confidence; this view is suspect. The technique is clearly an agency issue, and motivated by short-termism.
The current privileged position of the investment management industry is not tenable. It is clear that explicit fiduciary responsibility should, and will be, extended to their activities. In some countries, it is also necessary to consider the role of investment consultants . Once more this is an area where fiduciary responsibility is necessary, in particular because in many cases these consultants are commingling the roles of independent (investment) advisor and fund manager. For an extensive discussion of the role of and problems with investment consultants see: “Investment Consultants and Institutional Corruption”.
The issues in the public debate are principally over costs, fees and transparency, with an increasing emphasis on liability; of fiduciary duty. As noted by John Kay in a recent interview in The Actuary, most asset allocation methods follow the techniques developed in elementary financial economics, and are short-term in nature. The work of Beebower, Brinson and Hood showed that most of a portfolio’s return volatility could be explained by asset allocation – subsequent empirical studies vary but most attribute 80% or more to asset allocation. At a daily or intra-day level, almost all performance will be causally due to changes in price. All that this is really telling us is that price performance dominates returns in the short-term. This leads in turn to ever more short-term activities – notable among these are the hedging strategies of methods such as liability driven investment, and dynamic asset allocation- which are dependent upon the short term and market liquidity. To this day, most performance attribution analysis is conducted in this framework. It is perhaps not surprising that these attribution methods run into difficulty when moving to multiple periods or the long-term.
The long-term, however, is a rather different story. The classic study of long-term performance is Dimson, Marsh and Staunton’s work, published as “Triumph of the Optimists”, but also reproduced by many others. In this study of returns since 1900, the real return of 5.5% p.a. achieved by UK equities consists of 4.8% dividend yield, 0.6% growth in dividend yield and just 0.1% change in price. This is as intuitively obvious as the earlier observation that the short-term is dominated by price performance. It also applies to bonds – a bond held to maturity will return the face coupon and principal, and variation will only arise from differences in the achieved reinvestment rates of coupons, which are small convexity effects.
It is clear that investment managers should be using fundamentally different techniques and resource dedication when managing for the short-term than when managing for the long-term. The short-term is predominantly a game against others in markets; it is endogenous to the market. The long-term, by contrast, is driven by the fundamentals of the company, its earnings and dividends, and its likelihood of survival. This is exogenous to the market. Dedicating 90% of our analytic resources to price performance, and the game against others, may be justified when that accounts for 90% or more of the total available, but clearly, it is not when it accounts for less than 2%.
The earlier distinction by liquidity source goes beyond contract versus market, or income versus price; it is production versus valuation. Reliance upon the contract is “patient” investment; it is long-term. It is also “productive capital”. We might equally describe this distinction as the difference between investment and speculation, without any pejorative overtone.
Another important distinction arises here. The empirically observed lowering of volatility with holding period is a reflection of the growing importance of income in total returns with holding horizon rather than some inchoate reversion to the mean in asset prices; it is, in fact, convergence to fundamentals. This also raises issues with market prices for valuation. The market consistency justification for the use of market prices is further weakened when it is observed that that they introduce the volatility of the short-term into valuation.
Markets are also not consistent among themselves; if they were, diversification would not be feasible. The contracts written by long-term insurers and pension schemes are not negotiable; they lack the option on market liquidity of traded negotiable instruments. It should be realised that liquidity is costly; if it were not, all assets would be liquid. The question is not which market should we require these institutions to be consistent with, but rather: why should we shoehorn these institutions into a framework where they are not naturally members at all?
Engagement (or commitment) is currently a popular concern, usually rooted in the principal agent problem between investors and managers. Since Hirschman, this is often expressed as the difference between ‘exit’ and ‘voice’. For many investors, notably short-term players, it may be more efficient to ‘exit’, by sale in a market, rather than to engage, since engagement is costly. For the long-term investor here, the effect of management ‘rents’ is directly visible, and engagement to rectify these issues more probable, as the trade-off can often then be measured: though quite how one would measure and evaluate management ‘shirking’ is difficult to see. The economic term ‘rent’ may need some explanation; in simple terms it is the difference between what was paid and what might have been paid under perfect competition.
One of the issues for the short-term investor is that the purpose of this engagement is to affect prices, from which all benefit, though most have not incurred engagement expense; this may place them at a competence disadvantage to other short-term players. This is the ‘free-rider’ problem. However, the long-term investor is unconcerned with market price performance; engagement is a matter of self-interest. Engagement, per se, is a healthy symptom rather than a defining characteristic of the long-term. This question of the differences between the short-term endogenous game against others and the long-term exogenous game against nature is itself important in terms of investment management incentives. In the former, players are concerned with their position relative to others and price performance matters only to the extent to which it attracts further savers to the market. There is no incentive for such short-term players to want prices to reflect fundamental value, and often many to prefer distortions of this relation. By contrast, the long-term investor should want to see improvements to the fundamental productive output, and be unconcerned by market price behaviour, and may even want them to decline – for example, when future investment inflows are large. This brings to mind Warren Buffett’s famous hamburger analogy – if we intend to consume a hamburger daily for the rest of our lives then we want the price to go down, not up (see also our brief earlier discussion of housing).
The form of distribution preferred also varies as we move from the short to the long-term. The long-term investor will typically prefer special dividends to share ‘buy-backs’.
Perhaps, the greatest paradox of the short and the long-term lies in the incentives for research and innovation, which is intrinsically (but differently) a speculative activity. The short-term player will discourage these investments in pursuit of the immediate, while the long-term investor will encourage them. Ultimately, of course, the benefits of innovations will accrue to consumers, but in the short-term, they are a source of supra normal profits to producers and investors. The far more important point, especially from the point of view of long-term investment or finance, is that it is research and innovation that are responsible for growth in productivity and productive output - something we have seen precious little of in recent years.
The question of incentives also applies in the realm of risk management. Both insurance companies and pension funds are subject to regimes of risk-based solvency tests. These are immediate, with asset valuation based on market prices, discouraging consideration of the long term.
One of the few things we know about risk is that it means that more things can happen than will. Holding current provisions against all that might occur introduces significant redundancies, which are costly. The assumption is that the portfolio will visit all future possible states, but the reality, when investing for the long-term, is that few of these will be visited at times when they are relevant.
It is clear that recent regulation has had significant effects upon the asset allocation strategies of both pension funds and investors. Hedging has grown greatly in popularity. It should be realised that the only perfect hedge of an asset is the sale of that asset; and that imperfect hedges, apart from requiring constant maintenance, may compound losses. This leaves the institution either perfectly hedged and holding only cash, which is hardly consistent with any concept of investment; or imperfectly hedged and subject to additional unexpected and unplanned events.
It is clear that the mandates awarded by pension funds and insurance companies will not change unless regulation is more accommodative. Until it becomes clear that this is likely to occur, it is unreasonable to expect fund managers to modify their business models and processes to long-term management. There is a further issue: in the absence of some substantial evidence of the superiority of long-term investment, the terms on which mandates are awarded and the ways in which performance is measured by investment institutions, and their advisors, may continue to frustrate the best intentions of investors, fund managers and the authorities.
The fund management industry has long enjoyed a privileged position – trust with extremely limited liability. Investment mandates owe much of their length and fine print to indemnification of fund managers against all but gross breaches of these contracts. Even the newer fund management technique, known as fiduciary management (and a range of other expressions such as implemented consulting), suffers this problem. The documentation is usually concerned extensively with manager indemnification.
In parallel, in the short versus long-term investment debate, institutional investors and their fund managers have been condemned as failing to satisfy the socioeconomic role of investment. The better analyses do indicate that risk-based regulation and inappropriate accounting standards have played a major part in encouraging the trend to short-termism, but that carries little weight in the popular view.
Transparency is rather more than disclosure of comprehensive costs. Fund management marketing materials owe much to the dark arts of advertising. If trust is to be rebuilt, ignoring the existence of risks or over-promoting benefits will not help. The existence of the possibility of loss is a required condition for trust to be necessary. For trust to be maintained and reinforced, it is critical that the likelihood and magnitude of those adverse outcomes should have been fairly described. Over-promising is injurious to the maintenance of trust.
It is notable that the crisis spawned many lawsuits between fund managers and institutional investors. Many have been settled, but few offer comprehensive descriptions of the terms of settlement. Some are even subject to super-injunction clauses that deny reference even to the existence of the settlement. How does this contribute to the development of trust and confidence?
The mandates should be awarded under terms that impose fiduciary responsibility on the agent fund manager. This would extend to all of the trustee responsibilities as to costs and performance in the fund management aspect of their duties. This may be taken as a mandate to minimise all costs, while maximising long-term performance within the risk constraints of the mandate. This fiduciary duty would require the fund manager to hold capital resources or indemnity insurance in respect of the potential liabilities. This would include liability for breach of mandate, including any which have occurred but not resulted in economic loss. The term of award should be long term; we would prefer these to be decades rather than the often-cited three years. We would prefer narrative reporting to performance analysis. The term of this reporting should be annual. With cause, an investor could always terminate a mandate made as a long-term award.
We do not believe that performance-based incentive fees for fund management are appropriate for long-term mandates. They encourage hyper-activity in trading and operate to the detriment of the long-term, and also of markets. There is empirical evidence that liquid markets converge to fundamentals, the holy grail of the long-term, more slowly than illiquid. The rationale here is that activity in liquid stocks is dominated by uninformed “noise” traders while the informed concentrate on illiquid stocks. Thus liquid stocks should be expected to bear a liquidity premium and greater volatility in relation to illiquid stocks, leading to a much less attractive Sharpe ratio.
We have focused here on institutional rather than retail fund management in large part because there are major revisions of practice underway in the retail sector. In addition, the UK Office of Fair Trading has an inquiry into fees and costs for pension savings under way.
The UK Law Commission has been tasked with looking at this issue. Its terms of reference are very broad. We reproduce them below:
Terms Of Reference
1.1. On 26 March 2013 the Law Commission received a reference from the Secretary of State for Business, Innovation and Skills in the following terms: To investigate the extent to which, under existing law, fiduciary duties apply to:
1. To investigate the extent to which, under existing law, fiduciary duties apply to:
a. Intermediaries (including investment managers and pension scheme trustees) investing on behalf of others;
b. Those providing advice or other services to those undertaking investment activity.
2. To evaluate what fiduciary duties permit or require such persons to consider when developing or discharging an investment strategy in the best interests of the ultimate beneficiaries: in particular, the extent to which fiduciaries may, or must, consider:
a. Factors relevant to long-term investment performance which might not have an immediate financial impact, including questions of sustainability or environmental and social impact;
b. Interests beyond the maximisation of financial return;
c. Generally prevailing ethical standards, and / or the ethical views of their beneficiaries, even where this may not be in the immediate financial interest of those beneficiaries;
3. To consult relevant stakeholders in the equity investment chain on their understanding of what the content and application of fiduciary duties in this context is, or should be, and to consider their responses;
4. To consider whether fiduciary duties, as established in law or as applied in practice, are conducive to investment strategies that are in the best interests of the ultimate beneficiaries. In particular, to consider whether the duties:
a. Reflect an appropriate understanding of the scope of beneficiaries’ best interests;
b. Are sufficiently certain in content and application to market participants;
c. Permit sufficient diversity of strategy;
d. Sufficiently encourage long-term investment strategy and consideration of factors which might impact on long-term investment performance;
e. Allow fiduciaries to invest in line with generally prevailing ethical standards, and / or the ethical views of their beneficiaries, even where this may not be in the immediate financial interest of those beneficiaries;
f. Require a sufficient balance of risk and benefit; and
g. Are sufficiently balanced against each other.
5. To identify any areas where changes to fiduciary duties are needed in relation to these criteria and to make recommendations.
These terms of reference are sufficiently broad that we prefer not to comment on the possible outcomes. However, we would like also to draw attention to two papers on this subject recently published in the Rotman International Journal of Pension Management. The first of these is Reclaiming Fiduciary Duty Balance by James Hawley, Keith Johnson, and Ed Waitzer. The abstract of that paper (reproduced below) provides a good summary of many of the issues:
Fundamental fiduciary principles have survived intact for centuries, yet their interpretation has been dynamic. Given changes over the past few decades in global economic, capital management, and market structures, we appear to be at another inflection point in the understanding of fiduciary principles. Years of focus on the fiduciary duty of prudence has generated myopic investment herding behaviors, undermined intergenerational pension equity, and disrupted attention to the fiduciary duties of loyalty and impartiality. Reclaiming fiduciary duty balance between prudence, loyalty and impartiality is critical to sustaining pension promises. It would encourage better alignment of pension service providers’ supply chain interests, adoption of fitfor- purpose pension fund governance practices, and implementation of precautionary risk management policies.
The second paper is Modernizing Pension Fund Legal Standards for the Twenty-First Century by Keith L. Johnson and Frank Jan de Graaf, which offers one vision of a possible way forward.
This paper argues for a modernized interpretation of fiduciary duty that recognizes the symbiotic relationship between the sustainable success of both corporations and pension funds. It describes the impact that pension investment practices have on both the well-being of fund participants and the health of the global economy. It also argues that fiduciaries should adopt pension fund governance practices found to be associated with improved investment performance, better align pension fund service provider incentives with the clients’ long-term interests, and expand risk identification and management practices to consider systemic and extra-financial factors that may not be reflected on corporate financial statements but have contributed to the current financial crisis. The authors recommend development of pension fund governance best practice guidelines combined with adoption of a ‘comply or explain’ reporting scheme as a way to improve the ability of pension managers to meet their fiduciary obligations and promote economic stability.
Some caution should be added to these views. It is far from obvious that interpretations of the law on fiduciary will alone be sufficient, and there are numerous conflicting objectives. To quote Roger McCormick, the head of the London School of Economics' Sustainable Finance Project:
“From time to time, you hear suggestions and ideas about fiduciary duty, which tend to point the law in different directions. You have the ESG [environmental, social and governance] community wanting to have a more liberal interpretation of what the duties are. Then you have other people looking at investment banks selling dodgy-looking derivatives products to people that they call 'muppets', and maybe feel that a higher level of duty should be imposed in situations like that. You have to break the questions down in a more analytical fashion, almost forget the label 'fiduciary duty' and focus on what the substance of the duty should be.
"You can't expect trustees to take risk in this area if the law isn't reasonably clear – they are not going to be interested in sticking their necks out.
"With that as a background, [that the Law Commission's review would allow for a considered, objective analysis of what the current interpretation of the law entailed] we can have a more enlightened debate on the political aspects – and there are a lot of them. This is all about what you do with other people's money – you can't get more political than that. It's very difficult to lay down a hard and fast rule that will work for decades to come.”
In response to the first question we noted: “The failure of the conventional and unconventional paradigms is in providing a compelling description of the corporation.” Clearly, non-financial information can help significantly in this context.
Integrating non-financials especially environmental, social and governance (ESG) considerations into investment decision-making enables investors, investment analysts and creditors to make an informed assessment of the ability of a company to create and sustain value over the short-, medium- and long-term.
A growing number of investors, in particular those large institutional investors committed to creating long-term value for their customers, look for a more complete picture of organisations. The rationale for incorporating non-financial aspects into investment decisions are the following:
As a consequence, it is in the self-interest of listed companies to provide high- quality non-financial reporting as it documents that the company understands its business and the relevant challenges and that it is effective in steering the company towards a sustainable value. This in turn attracts investor interest and may raise liquidity in the company’s own stock.
On 16 Apr 2013 the European Commission adopted a proposal for a directive enhancing the transparency of especially large companies on social and environmental matters as an amendment to Accounting Directives (Fourth and Seventh Accounting Directives on Annual and Consolidated Accounts, 78/660/EEC and 83/349/EEC, respectively). With this proposal the European Commission states as an objective to increase transparency and performance on environmental and social matters of companies in the EU.
In as much as EFFAS welcomes the proposal of the European Commission, we would like to caution expectations about the impact of the amendment to the Accounting Directives.
For non-financial information to be useful for investors it has to comply with certain criteria of quality and relevance; notably comparability, quantifiability, materiality in terms of risk/opportunities, and being in appropriate, i.e. non-prose, format.
Today, only a few reporting frameworks take into account investors' information requirements vis-a-vis non-financial reporting. Also, only a few of the current national legislative approaches for non-financial disclosure contain concrete guidance on which non-financial aspects should be disclosed and in which format.
No. We regard the introduction of benchmarks as an invitation to adopt replication and information-free strategies and with those, short-term approaches to portfolio management. We are particularly concerned by issues that arise with “synthetic” financial contracts that may be based upon benchmark indices. Here a financial institution may write a contract with its performance is linked to the index performance. If we were to consider a single security, we may see that this amounts to increasing the supply of that security; in this, it shares some aspects with short selling. It does not act, though, directly on the traded price of the security. However, it might be argued that it has reduced the market demand and perhaps, market price of the security; as such, it might be considered a form of manipulation. Often these synthetic contracts are structured to involve minimal cash payments at their inception – these are, by design, speculative instruments. This can also be an issue with some Exchange Traded Funds.
There can also be issues of taxation involved. This was the heart of the “manufactured dividends” concerns, which were, for example, prevalent in the UK gilt market prior to its modernisation in the early 1990s, when short selling was first permitted.
We follow the Eurostat definitions of SMEs in our responses to these questions regarding SME finance. So, an SME must have less than 250 employees and a turnover of less than €50 or balance sheet of less than €43 million. The UK classifies businesses with turnover up to £25 million as SMEs and those with £25 - £500 million as “mid-size”.
There is considerable variation across Europe in the role of SMEs in investment, as may be judged from the illustration below.
There is also considerable disparity among European states in the role and composition of SMEs. This is illustrated by the variation in turnover across the SME industrial classification sectors.
Though SMEs may not have access to equity capital markets, they do benefit from a broad spectrum of sources of finance. Though 80% of SME finance is bank sourced, it is easy to overstate the reliance of the sector upon banks. According to Eurostat, 36.1% of SMEs have never required a bank loan, 39.9% have never used a bank loan or overdraft and 60.9% have never sought grants or subsidised loans. The diagram below illustrates the sources of SME finance for France and the UK.
The credit crisis has undoubtedly affected the availability of bank finance to SMEs. This can be judged by comparison of the rate of loan application refusal in various European countries.
Contrary to popular opinion, though, the terms of bank loans have not hardened markedly everywhere, as is illustrated below.
The European Central Bank conducts a survey every six months of credit condition in the Euro area. This most recent shows similar mixed results and is reproduced below.
The ECB survey goes further and asks SMEs for their most pressing problems in terms of business objectives. The response that “Finding Customers” was the most important is most interesting as in our experience, companies will invest when they see clear demand for the products. This survey merits close reading.
And, in response to the crisis and the perceptions of greater difficulty for SMEs to finance themselves, we have seen a welter of official interventions from which SMEs may benefit, which are listed below.
It is not at all obvious that these interventions have yet had sufficient time to be fully effective.
SMEs benefit to varying degrees from state sponsored sources of loans and guarantees across Europe.
We have, in other responses, indicated that the use by corporations of insured book-reserve pensions would bolster their finances and reduce dependency upon the banking sector. As the amount of notional contributions to these schemes is directly related to the employed workforce, this would differentially provide finance to SMEs. The proportions of the labour force employed in companies of different sizes are shown below.
That said, we have been unable to determine the importance of SMEs in long-term investment. As many are young companies and the rate of dissolution of new ventures is high (for example, in the UK 50% of newly incorporated ventures do not survive their fifth birthday), we suspect that it may be far lower than that of established large enterprises.
Following the recommendations of the Breedon Report, the UK Business Bank commissioned a study from AFME (Association of Financial Markets in Europe) on the securitisation of SME obligations. This paper, “An Agency for Business Lending – “ABLe”-Improving access to finance for small and medium-sized enterprises”, discusses many of the issues with SMEs and securitisation and proposes in outline the creation of an agency to facilitate the securitisation of SME obligations. The paper refers to and recommends reading an accompanying “Technical Paper” for greater detail. AFME have declined to supply us with a copy.
We have numerous and substantial reservations with the approaches proposed in this paper. These cover aspects, such as guarantees and special regulatory treatments for such lending; we are reminded of the Deutsche Wagnisfinanzierungsgesellschatt and cautioned by that experience. We will withhold further comment now, as we have not seen the detail of the Technical Paper that may possibly address these. We have made a request, under the Freedom of Information Act, to the UK Business Bank for a copy of the Paper. We are frankly appalled by the lack of transparency that is evident here; this will hardly help in developing trust and confidence in a new putative new institution and market.
We are concerned that with more debt finance available to SMEs, transactions will occur that would not have under the current model; this is the heart of additionality. Our concern is, in essence, that many loans will be made that prove irrecoverable. This extended borrowing population would render the already poor data for SME performance unreliable and the securitisation pools much riskier. Here, we are fishing is a population where the better fish have already been removed. If, on the other hand, lending to SMEs is not higher, then the securitisation process reduces to a hidden subsidy for the bank originators.
The private information aspect of the performance of SME loans leads to a co-ordination problem, which means that the banks should not be expected to develop common credit registers among themselves. The absence of such registers, and detailed data on the past performance of SMEs, makes the introduction of non-bank lenders into this market more difficult. There is a potential role here. Ensuring the participation of banks in submitting data could be achieved by restricting access of non-participating banks to other facilities such as credit enhancements and state sponsored loans. However, we are unconvinced that further actions to enable long-term investment are warranted at this time.
In the past few years in some EU markets, such as Spain, Austria, Germany and the UK, midsized companies, often unlisted, have begun to issue corporate bonds. A variety of methods of issuance have been used: stock exchange listing and OTC.
The majority of mid-sized companies in Europe, especially in established industries, are able to maintain growth through internally generated cash-flows. The motivation of the majority of midsized bond issuers, however, was to finance their going concern needs; to finance growth of the ongoing business. As was noted earlier, equity markets seem no longer to fulfill this role even for large companies.
We are concerned that the current development in some of the newly created mid-sized bond markets already show symptoms well known from the new ecomomy bubble at the end of the last century. In response to this, EFFAS has published "Minimum Standards for Bond Communication" which define how corporate bond issuers should communicate the terms and conditions, as well as covenants of their bond to safeguard transparency and fairness of their offer. This is particularly relevant given the high retail participation in these markets.
As for securitisation of SME finance, the primary issue is that of moral hazard. We refer the reader back to our answer to Question 14. We would note though that the likelihood of default for SMEs is usually estimated to be higher and more variable over the business cycle than for either consumer lending or corporate lending more generally. Moreover, the profitability of the SME segment is usually lower than that of corporate lending or the retail segment.
Given the high levels of personal guarantees from owners and officers in SME lending, it is also clear that the securitisation process would need to contain a method of shielding investors from the post-insolvency issues. A guarantee of, or minimum assured recovery, perhaps.
SME loans are not an obvious candidate for the application of securitisation techniques. Given the information asymmetries and lack of knowledge of investors with respect to the companies in a securitisation pool, we wonder if this would not be a more expensive finance than traditional bank loans.
The role of SMEs in European economies varies considerably from state to state as may be judged from their relative shares is economic value added. In Germany, SMEs account for 53% of GVA, in France 56%, in the UK 50%, in Spain 68% and in Italy 71%.
Some very different approaches have recently been taken to the question of SME finance, notably the UK Department of Business Innovation and Skills “Business Bank”. This bank arose from the Breedon Report that considered lending channels in the UK; it addressed the question: “In light of the need to find new sources of capital to support the future financing needs for UK businesses, what are the barriers to the development and growth of sustainable non-bank lending channels in the UK?” This has total capital of £3.9 billion, of which £2.9 billion was already allocated under pre-existing programmes. Its objectives are to:
Its April 2013 update the range of existing SME support schemes in the UK. This is reproduced below.
Notwithstanding, these schemes and the Business Bank which is not yet fully operational, this update identifies the following as underserved by finance in the UK:
It is not at all obvious that markets are well suited to address any of these issues. There are issues of scale and fungibility. Perhaps, the nearest that we can come to a market solution would be securitisation of loans made by a European equivalent of the Business Bank or the issuance of subordinated debt or loan guarantees in a manner similar to the EU Project Bonds.
We do not believe that regulation is an appropriate means of address. Much of the finance under consideration here is offered by entities that are regulated institutions in other contexts, such as insurance companies and pension funds. Other sources are specialist institutions, such as leasing companies and investment funds. We do not believe that it is appropriate to regulate such institutions if they are not in the business of taking deposits and are not systemically important. We believe that the best way to achieve continuing growth of the non-bank financing of SMEs is to do nothing. The best we can suggest is to separate the questions of SME finance from questions of long-term finance and infrastructure and sustainable investment.
If the objective is to increase non-bank lending, the approach should be one of encouragement not imposition. For example, it might be possible to require insurance companies to hold some minimum fixed proportion of assets in SME advances, but that would merely distort two markets. This issue has been greatly confused by attitudes to ‘shadow banking’ and the possible regulation of these activities.
The creation of a credit register will go far in reducing the greatest impediment to increased lending bu non-bank institutions.
Many of our responses have been concerned with the creation of additional incentives for long-term and sustainable investment. We would be concerned that incentives offered did not overcompensate and reduce the levels of other productive shorter-term investments. The ‘crowding out’ mentioned in our response to question one, is not just of other forms of investment but also within the term structure of infrastructure and sustainable investment.
With this in mind, we note that saving and long-term investment is being promoted without any recitation or analysis of the superiority of this to, for example, repeated sequences of short-term investments. We would suggest that the EC commission academic studies which elucidate these question: Why is long-term investment superior to the short term?
We can see that taking a short-term view may result in some projects not being undertaken, as they may be unprofitable in the short term (a classic J-curve) though of positive net present value over the long term. We are aware that growth models will typically produce higher returns in the long-term than as repeated sequences of short-terms. It is also obvious that uncertainty may result in long-term projects that are viable not being undertaken, or indeed being abandoned in midprocess. There are then many other related issues, such as positive externalities. In this regard, we are particularly concerned that many government-sponsored projects are often based upon projections that are questionable and often inadequately researched; these projects are often driven more by political will than by economic reality. These are unlikely to survive the private sector’s due diligence processes. Even if they did, the outcomes for investors are then likely to disappoint and damage trust in this form of investment. The UK National Audit Office’s “High Speed 2: A review of early programme preparation.” provides an illustration of some of the issues in an infrastructure context: Amyas Morse, the head of the National Audit Office, said:
“It’s too early in the High Speed 2 programme to conclude on the likelihood of its achieving value for money. Our concern at this point is the lack of clarity around the Department’s objectives.
“The strategic case for the network should be better developed at this stage of the programme. It is intended to demonstrate the need for the line but so far presents limited evidence on forecast passenger demand and expected capacity shortages on existing lines. It is also unclear how High Speed 2 will transform regional economies by delivering jobs and growth.
“The Department is trying against a challenging timetable to strengthen its evidence and analysis, which at present provide a weak foundation for securing and demonstrating success in the programme in future."
We feel that this recommended course of action would be warranted on several grounds.
Firstly, the culture of the short-term and current practices are deeply embedded and investors will need to see benefit from the long-term to change. Change is usually costly. However too much of the discussion seems to be based on the idea that long-term investment is virtuous – to be regarded in much the same way as Victorians regarded thrift, as a merit in itself; that is clearly not sufficient.
Secondly and perhaps more importantly, tax incentives will not work for long-term projects if those projects are seen to be political in nature and not financially self-sustaining. The time inconsistency of political processes is well understood by the investment community. Finally, there is the question of the cost of these incentives. If these long-term projects are, intrinsically, more rewarding, then the levels of incentives may be lower and might even be phased out entirely over time.
With this cost of incentives element in mind, we were a little surprised by the absence of substantial reference to behavioural science and its applications in finance. This would appear to hold the prospect of achieving much in terms of cultural change at little cost, and perhaps, might remove the need for some incentives.