The European Pension Debate: How Pension Fund Deficits Affect Credit Ratings

The European pensions debate has been in progress for some time, although little clarity has been established from a credit analysis point of view. The aim of this paper is to provide insight into the pension accounting framework surrounding defined benefit schemes in five major European markets, whilst offering a range of qualitative and quantitative considerations relevant in performing credit analysis on pension issues.

There is a common misconception that the debate has been fuelled by the decline in the equity markets. Whilst this is true for the United Kingdom, where the schemes are specifically invested in these types of assets, this is not necessarily the case for the rest of Europe, where pension funds are provided on the companies’ balance sheets rather than being invested in specific assets that carry their own risk characteristics.

The general rise in pension liabilities does require some modification to Fitch’s financial ratio analysis. Fitch will start to add pension fund deficits to financial indebtedness as a supplementary tool in our quantitative financial analysis and compare this with a company’s profitability and cash flow generating ability. For funded schemes the whole of the deficit will be added and in countries where pension fund liabilities are carried as a balance sheet provision, an estimate of the proportion of the liability funded by debt will be added. Fitch’s coverage ratios will also be adapted to establish the ability of a company’s profits and cash flow to cover pension contributions and the imputed interest cost of funding its pension liabilities.

These ratios will not become the leading part of our analysis, but will enable Fitch to quantify more accurately the effect that pension fund deficits have on credit ratings and thereby improve transparency for the investor. Pension liabilities are long-term in nature and the degree of liability has been currently heightened by the decline in equity markets, although market recoveries may mitigate this. Adding pension deficits to other long-term liabilities such as financial indebtedness gives an appropriate financial means of measuring pension liabilities.

Our initial conclusions are that European corporate credit ratings are not changing in a wholesale fashion because of this issue. The current size of the decline in pension fund assets is more likely to generate rating outlook changes for those companies most affected rather than rating changes. It is possible that a few ratings may be lowered by around one notch. Any such changes would be implemented primarily as a result of the 2002 year-end earnings announcements, the full details of which will become available over the next one to two months or so. Fitch believes that the fundamental business and credit profile, and the performance of a company will continue to be the main determinants of its rating.

Convergence of Accounting Standards

As the implementation of International Accounting Standards (IAS) for European companies is approached (expected 2005), it is clear that significant additional convergence needs to take place in order to allow true cross border analysis of companies and in particular pension arrangements for employees. These arrangements have traditionally grown out of the political, economic and cultural landscape and tend to be particular to each country as a result. In the United Kingdom, the current transition to FRS 17 has provided additional transparency with respect to pension funds, although the financial accounts of the majority of companies still reflect the old reporting requirements.

One of the side effects of the proposed introduction of the new style of pensions accounting and disclosure in the UK has been the closure of a number of defined benefit schemes, with the employers effectively outsourcing the resulting FRS 17 balance sheet volatility to the employees. Fitch has conducted a survey of its rated European corporates. Some 39% of the schemes reviewed in the UK had been closed with others introducing new measures, e.g. lifetime average pensions, eligibility after a minimum service period of for example five years. This compares with the 33% of the FT 350 companies (from a Mercer survey), which have closed defined benefit pension schemes.

The debate surrounding the treatment of defined benefit pension schemes has intensified over the past year following the poor performance of equity markets, affecting both the size of the schemes’ assets and also the size of the liabilities relative to the companies’ market capitalisation. This, in conjunction with the harmonisation of European accounting standards to IAS in 2005, has created significant interest in the matter.

What is clear from this analysis is that significant differences exist in the structures used to provide defined benefit pensions across the countries selected. The adoption of a common standard to account for these obligations will in part aid credit analysts in providing an equivalent level of disclosure. However, it is likely that the underlying investment strategy for the schemes will remain specific to each nation, which may result in adjustments of varying degrees of magnitude. It is also possible as part of the ongoing harmonisation across the European Union (EU) that pension creditors will be granted simultaneously the status of preferred creditor in those jurisdictions where this is not currently the case.

United Kingdom

In the UK significant debate has surrounded the phased introduction of FRS 17, the new pensions accounting standard designed to eventually replace SSAP 24. The International Accounting Standards Board has announced a review of IAS 19 in view of the forthcoming convergence of accounting standards across Europe. Fitch believes that there will be few changes since US GAAP and the current IAS (IAS 19) are similar in many respects to FRS 17.

The UK Accounting Standards Board has now proposed to defer the mandatory adoption of FRS 17 during the consultation on IAS 19, and then to consult on the early adoption in the UK of the standard that the International Accounting Standards Board puts forward. As a result FRS 17 looks increasingly unlikely to be adopted in its current form. However, during this period of uncertainty UK accounts will need to include disclosure of information prepared under the FRS 17 transitional arrangements.

The table overleaf shows the key differences between the two UK standards. The key difference is in the approach, with SSAP 24 spreading any deficit over the average remaining service lives of the scheme members, using the ‘matching’ principle according to a conservative actuarial valuation methodology. FRS 17 requires a mark to market approach, applying the principle of prudence since if the company was to be wound up today that would be the amount available in the pension fund. With the increased disclosure that companies now provide under FRS 17 (even at this early stage of implementation) more insight into the schemes and risks involved can be achieved.

In view of the traditional provision of defined benefit schemes to employees in the UK this has translated into a key area of discussion. The Treasury and DSS have published the government’s proposed statutory replacement for the minimum funding rules for pension schemes, in which they put forward that any shortfall in the fund on winding up the company should be a creditor of the company potentially with preferential status. Such a step would subordinate the unsecured creditors and as a result the importance of the pension fund assessment would increase even further in credit analysis in the UK. Upon liquidation the pension creditor would take priority, affecting the recovery prospects of the senior unsecured creditor.

Currently, if the company breaches its MFR (Minimum Funding Requirement) it may be required by the pension fund to grant a charge over certain assets to make good the deficit. However, this is the only way that the pension fund can assert its claim before that of the senior unsecured creditors. Unpaid pension fund contributions do attract some limited preference upon insolvency, however, the extent of such claims is limited to the unpaid contributions. The pension trustees have a fiduciary duty to ensure that these contributions are made on a timely basis.

Fitch focuses on the following areas as important in both the quantitative and qualitative assessment of the company:

• Whether the scheme is defined benefit or defined contribution.
• Distribution of investments between equity and other investments, e.g. bonds, gilts, cash and property.
• Split of the fund between current employees, deferred pensioners and current pensioners, and whether the scheme is still open to new members.
• Assumed rates of return and discount and other rates applied.
• Size of the FRS 17 deficit relative to the net assets of the company.
• Recoverability of any surplus.
• Requirement to make good any deficit

Pension Fund Split Sorted by Asset Size

Distribution of Investments and Profile of Members

The distribution of investments is a crucial element in the appraisal of pension liabilities from a credit point of view. This is a qualitative issue and is closely related to the distribution of members and whether the scheme remains open to new joiners. If the scheme is full of young workers then it may be appropriate to invest 100% in equities, in view of the greater investment returns expected over the working lives of those employees.

If the majority of members are already retired, and the scheme is closed to new members, then it may be more appropriate to invest predominantly in bonds since the outgoings from the fund require some degree of certainty around the income stream.

Split of Fund by Member Type

Assumed Rates of Return and Valuation Assumptions

Fitch observed that there was some variation in the assumed rates of return relative to the discount rate applied. The standard states that the rate used to discount the liabilities should be equivalent to that of a ‘AA’ corporate bond. This is because by applying a differing discount rate for the assets and liabilities a surplus or deficit can be created or artificially widened or reduced (see appendix for graphs).

It is widely believed that this may lead to a number of companies switching their pension funds into ‘AA’ rated corporate bonds to match the assets and liabilities in some way.

The study above suggests that the move into a 100% bond portfolio could still result in a cash cost to the company in terms of increased pension contributions to fund the deficit created over time, whilst increased returns may be found for these funds elsewhere within the company.

In fact FRS 17 creates a vicious circle in an economic downturn, whereby predominantly equity invested funds will fall sharply, whilst the rate of return used to discount the liabilities also declines and the ability of the individual company to pay increased contributions decreases. If pension fund members were to become preferential creditors this would become an area of quantitative modelling for the credit industry.

Size of the FRS 17 Deficit Relative to the Net and Total Assets of the Company Fitch compared the size of the deficit (defined as FRS 17 deficit/surplus including tax credit) to the net and total assets of the company (valued at the most recent balance sheet date). This gives an appreciation of how material the issue is for the company in question. If the deficit is greater than 50% of net assets it is crucial to consider the assets the fund is invested in and the profile of the members as well as whether the scheme is still open to new members. Fitch selected net assets as a comparator rather than market capitalisation since, generally speaking, the funding deficit will increase/decrease with the market and hence the market capitalisation, whilst net assets will remain a more constant measure of size. Balance sheet net assets also represent the buffer which protects creditors against loss. Although the accounting policies adopted to value assets can be debated, using this measure does give a more permanent measure of the scale of pension fund problems an entity may have.

The survey sample in October 2002 showed that only one company in the survey had a deficit in excess of 50%, whilst a number of companies had negative net assets or small surpluses.

FRS 17 Deficit (Incl Tax Credit) to Net Assets Sorted by Total Asset Size (Most Recent Year End)

Fitch also compared the size of the deficit to the size of the scheme and the size of the scheme to the total assets of the company, to gain an appreciation of the materiality of the issue. Whilst it can be seen that the deficit is generally below 20% of the size of the scheme, it is notable that the size of the schemes generally represent a higher proportion of total assets and also of net assets as shown in the graph above. If these metrics start to increase consideration should be given to the quality of the assets, and also the assumptions used in calculating them, since a small percentage difference in the rates of return or discount rates can have a material effect upon the current valuation of the fund and any deficit or surplus.

Recoverability of Any Surplus

A FRS 17 surplus may only be taken into account when considering the credit profile of the company if it is recoverable. Generally, there are a number of formal steps that need to be followed before a surplus can be returned to the company. These steps will vary from company to company, depending on the construction of the pension trust. Fitch would therefore assess the likelihood of the recovery of the surplus, based upon discussions with the company and legal opinion on the matter, prior to incorporating the surplus into its analysis.

Relationship Between Size of Scheme to Total Assets and Size of Deficit to Total Scheme Size

Requirement to Make Good Any Deficit

Fitch noted that some survey responses indicated that companies had insured out the risk of their pension funds not meeting the liabilities over time, hence removing any potential funding gap. This is taken into account as part of the qualitative analysis by the agency. It will clearly largely eliminate the risk of any rating downgrade. Fitch will monitor the approach that rated corporates take in this particular aspect of the pensions question as they outsource the investment risk to the insurer.

Of note, one company has recently started to change its final salary schemes. Whilst keeping the defined benefit type scheme in place, the percentage of final salaries, which employees will be entitled to upon retirement, has been reduced in order to reduce funding gaps.

Conclusion – Credit Treatment of UK Pension Funds

The foregoing commentary highlights some of the analytical steps Fitch will consider in assessing the impact of pension fund deficits on credit ratings. The principal determinant of ratings will remain the business and financial performance of a company. A well performing company with strong profitability and cash flows should be able to handle pension fund deficits with relative ease. For example, British Telecom has stabilised its financial profile during 2003/2004. It is increasing substantially its annual contributions to its pension fund, but the more important stabilisation, which has occurred has led Fitch to revise its rating outlook to stable from negative.

Having said this, Fitch will assess the quantitative impact of a pension fund deficit on a corporate by adding that deficit to its financial indebtedness. The existing ratios, which quantify the financial leverage of a company, will remain important components of our quantitative analysis. Additional ratios will be calculated, adding a pension fund deficit to net debt and comparing that to the cash flow generating ability of the company. Pension fund liabilities are a long-term financial issue for a company, which will not be resolved in a few months and may even be mitigated to some extent by market recoveries over time. Proper consideration can be given to the degree of problems raised by a deficit by this means. A deficit may not have a large impact on this ratio. But, where it does, Fitch will be able to consider the company’s response to this issue more formally and whether any rating changes are required.

Further additional ratios will be calculated to measure the impact on the debt coverage ability of corporates. Fitch will adapt the standard EBITDA/interest ratios by adding pension contributions and the imputed interest cost of servicing the liability to interest expenses. The ratios adjusted will be EBITDA/net interest and EBITDA/gross interest. In the former ratio the net difference between the interest earnings on pension assets and the interest costs of pension liabilities will be added to the ratio’s denominator. In the latter, only the gross interest cost of pension liabilities will be added to the denominator.

In some instances companies are publishing the imputed interest cost and interest earnings. Where this is not occurring, Fitch will apply the rates used to discount liabilities and assets to the respective pension assets and liabilities totals, to obtain assumed values. The latter course is not ideal, but represents the only realistic step available given imperfect disclosure.

The adjusted coverage ratios will not have the same degree of effect on our ratings as the adjusted leverage measure. The servicing of pension liabilities does not represent a fixed charge on a business in the same way that interest expenses do, and they will vary over time. Hence, the effect of changing pension contributions has less of a quantitative impact on our ratings than our leverage calculations, but it will nevertheless be brought into account. (See Table 4 for worked example).

Rating changes are most likely where a company has to increase significantly the resources it is putting into a pension scheme and diverting funds away from the lenders to the company on a sustained basis. A long-term rating is not designed to be volatile and respond to a changing pension fund profile each year. Rating changes, as a result of this issue, are therefore likely to occur where lenders find the resources available to them are diminished in the long term.

Germany

To reach an opinion on how to deal with pension liabilities from a credit analysis point of view it is first important to determine who bears the liability. Does it arise by way of a direct liability to the employer? Or is it supported by external pension assets (which represent a portfolio of investments, which will fund future cash payments to the retired employee)? By far the most common form of pension obligation involves the former method, i.e. a direct obligation of the employers. Lately, larger German corporations have started to set up external pension funds in order to match Anglo-Saxon pension funding and accounting.

A major difference between UK and German pension accruals is that in the UK the funds are generally invested in a variety of financial instruments, whilst in Germany the funds corresponding with the provision raised sit on the company’s balance sheet and are in effect used as an internal source of funding (the accruals method). This, in turn, affects the cash based ratios used in credit analysis, as funds can appear under cash and quoted investments (if not otherwise used) under German GAAP. In addition, German pension accruals are effectively invested in the company itself and are dependent upon that company’s ability to pay in the future. A side issue of the unfunded status of German pension obligations is that there is also no formal reporting on the value of the assets corresponding with the pension provisions raised. Unlike US pension liabilities there is also no healthcare cover for pensioners included in German pension obligations. These are insured by separate social insurance that the beneficiaries have to join.

German pension schemes have to be a member of the Pensionssicherungsverein (a mutually-owned association that reinsures the pension obligations of the members’ schemes), which steps in upon insolvency, thereby protecting the pensioners. In Germany, pension creditors rank pari passu with ordinary unsecured creditors in an insolvency situation.

From a credit analysis point of view some adjustment needs to be made to take into account that the pension accruals method requires no direct funding on an ongoing basis. On a qualitative level, it is important to understand the materiality of the scheme in relation to the size of the company (defined as net assets), the distribution of members under the scheme, split between current pensioners, deferred pensioners and current employees to determine the maturity of the scheme. In addition, it is useful to know whether the scheme is still open to new members. Companies are switching, albeit gradually, to defined contribution funds, placing the risk on the employees.

There is a risk of a variation of assumptions between companies since the German Commercial Code (HGB) caps all discounts for pension liabilities to a maximum of 6%, although some conservative companies use discount rates of, as low as, 3.5%. However, the Code excludes future payroll increases and does not make provisions for employees younger than 30. German corporations have traditionally overstated pension provisions in an effort to reduce tax payments, and by overstating the pension obligations German corporations have probably made up for the payroll increases not being taken into account. However, this overstatement has only been used by the more profitable companies. Fitch thus views the level of discount rate applied as a measure of the conservatism of a corporation’s financial policy.

The agency has also noted that balance sheets show pension provisions for all pension obligations entered into on a global basis, which are subject to different legal regimes. It is therefore important for the agency to understand the location of the workers in order to assess the level of obligations that the group has entered into in different countries. While pension obligations are unfunded in Germany, they might be funded elsewhere and this fund represents the first port of call if the employee retires. Pension obligations can be held at different levels in German corporate groups. It is worth mentioning that Fitch will not adjust the value of the pension obligations according to the level at which these obligations occur within a group structure. Although they can, for example, be located at the holding company level – which can be the case after merging two groups, where the newly formed group holding company takes over parts of the existing pension obligations of the merged entities – Fitch does not consider these obligations to be subordinated in the same way that debt sometimes is at the holding company level.

Conclusion – Credit Treatment of German Pension Funds

Historically, Fitch has informally included German corporate pension liabilities in its analyses, considering the impact on the debt burden of the company. Our revised approach will incorporate these liabilities more formally and in a similar way to that for the UK.

The leverage ratio, however, will be adjusted by adding to the net debt total the proportion of balance sheet pension provisions, which are funded by debt. This will be compared with cash flow generation to measure the effective increase in financial leverage. This approach is underpinned by the knowledge that balance sheet pension provisions are highly unlikely to be required to be funded in full in a short time period. Therefore, the general financial parameters by which a company is funded can be used, with the pension provisions being assumed to be funded by both debt and equity. Although this is a fairly general assumption, it is in Fitch’s view the best way to capture the additional financial indebtedness incurred by balance sheet pension funding.

The coverage ratio will also be adjusted. The denominator of the EBITDA/net interest ratio will have the imputed interest cost of funding pension liabilities added. Accounting disclosure on this issue is quite weak. Fitch will therefore calculate the average cost of a company’s debt and this interest rate will be applied against the amount of debt added to net debt in the adjusted leverage ratio. The examples given show exactly how our ratio analysis will be adapted for off and on-balance sheet pension liabilities. (See Table 4 for worked example). This approach will be refined by analysis of the country of origin of the pension liabilities and their regulations. It is also important to understand a company’s approach to reducing a deficit.

Italy

By law, Italian companies must pay 7.41% of the employee’s annual gross salary as health and state pension social security costs. They must also accrue the TFR (Trattamento di Fine Rapporto), which represents an annual charge of about one month’s salary for each year of service. It is held as a balance sheet liability and must be paid as a lump sum, the payment of which is mandatory once the labour relationship ends, independently from the cause of termination. These accruals are not required to be invested in assets and hence have no real cash cost until the labour relationship ends. There is a statutory uplift in value each year, based upon a standard 1.5% uplift and an inflation related increase to ensure that the sums do not dwindle over time. However, in previous periods of high inflation this uplift methodology did not ensure that the liability increased at the same rate as prices. When employment ends the employee is given the full accrued amount.