Friday, November 7, 2014

Factors Influencing the Debt of Public-Private Partnership Projects in Italy

The Public-Private Partnership is by no means a new concept. Its origins can be found in early government concessions in China, granted for salt and iron mining about 2,000 years ago, according to the author, Gernet (1982). However, the phenomenon was only formalized in as late as the 18th century in France by the Public Works Concession (Concession de Travaux Publiques). Subsequently, it became a turn-to solution for large infrastructure projects for several succeeding governments in Western Europe, America and Asia.

However, despite the relative popularity of the Public-Private Partnerships, their arrival in Italy has been relatively recent. Rossi and Civitillo (2013) contend that these projects were first formalized in Italy in 1994 through what has become known as the Merloni Law. Since then as Figure 1 below illustrates, the popularity of Public-Private Partnerships has been on a general upward trend, both in terms of the number or projects undertaken and the collective annual outlay on these projects.

Figure 1. Italian Public-Private Partnerships, 2002-2012
Source: As noted on graph.

As noted above, Public-Private Partnerships, despite their ongoing popularity as a means of financing public infrastructure projects, are a relatively late arrival to the scene in Italy, having existed elsewhere for several decades. The reasons for the late arrival of the concept in Italy are difficult to ascertain. It is possible that one factor in the 20th century at least, was the frequency at which governments changed in Italy during the same century but the cost of government debt undoubtedly also played a role.

Figure 2 below depicts the trend of the cost of Italian debt since 1950. For the majority of the second half of the 20th century in Italy, real interest rates (that is, interest after inflation) were negative, effectively providing an incentive for successive governments to invest in public infrastructure rather than issue private contracts to do so. However, as the cost of issuing debt reached its apex in 1994 at the time that Public Private Partnerships were beginning in Italy, it’s reasonable to make an assumption that the cost of debt was a major factor here.

Figure 2. Italian Interest Rates, 1950-2012

Source: Banco d’Italia

In fact, the cost of debt is of the utmost relevance to this paper. Most Public-Private Partnerships are only around 10-15% funded by the SPV, which leaves a large proportion of the money to be made up by financial institutions such as banks – the focus of this paper. In particular, the paper will analyse how debt issued by banks (especially senior debt) is affected by the following variables:
-          Size of the Main Partner
-          Length of concession period
-          Interest rates on date of financial close
-          Solidity of the Main Partner
The paper will thereafter make a number of reasonable assumptions about how these variables usually affect the provision of debt for the Public Private Partnership.

At the outset, it is worth noting that the scale of the projects undertaken by Public-Private Partnerships are generally large enough to attract the attention of international financial institutions and not just Italian banks and financiers. This is also one of the tenets of the European Union - that all public projects are put to tender on an EU-wide basis. In theory at least, this ensures competition and transparency in the union and arguably creating a more efficient marketplace for infrastructure projects.

The implications for the EU-wide tender go beyond competition, efficiency and transparency, however. From an Italian perspective, it means that, on the government side (the “Public” in the Public-Private” equation), the Italian government at the time of the project in question is effectively competing with other countries’ infrastructure projects. In other words, when the NHS is looking for VFM (“value for money”) in its projects in the UK (Broadbent and Laughlin, 2003), this too has consequences for projects seeking financing in Italy.

Regarding the financial aspect of Public-Private Projects, Rossi and Civitillo (2013) note, “the funding of public-private partnership projects in Italy is generally granted by banks and rarely by capital market by selling bonds or shares to investors. Using such a kind of funding gives disadvantages in comparison with other countries: the interest rate is about 10-11%, while in the UK, for instance, the spread on the risk-free rate is about 0.75 – 1%.” The authors also note that Italian banks tend to ask for traditional guarantees for the financing.

The authors go on to note that the Italian government doesn’t use VFM as the UK government prescribes, but rather an “Economic-Financial Plan” – a financial model put together by the private side. This suggests that the process isn’t as transparent as it should be in Italy and that considerations beyond the size of the tender bid by the private (financing side) are potentially being given too much weight but the issue of transparency is not within the scope of this paper.

The paper will begin with a discussion surrounding some of the main theories of the use of debt in project financing – with a focus on Public-Private Partnerships. This is an area of the academic field which is still relatively new but is growing in line with the industry itself. This will be followed by analysis of what role the variables mentioned above (size, solidity, duration of concession period and interest rates) play in the issuance of debt for Public-Private Partnerships in Italy.

The Role of Banks in Italian Public-Private Partnerships
Banking institutions are the predominant form of finance for Public-Private Partnerships in Italy. It is somewhat difficult to ascertain why this is the case, given that the Italian debt market is the third largest of its kind in the world.[1] This could be related to the flexibility of bank financing relative to that of bond financing, its low cost and importantly to some SPVs, its low disclosure requirements (whereas Bond financing requires a public listing and much higher disclosure).

The predominant use of bank financing for Italian Public-Private Partnerships at least means, that for the purpose of research, the form of financing Italian Public-Private Partnerships can practically be simplified to banking institutions. Akintoye and Chinyio (2001) suggest the following structure for PFI (Public Finance Initiatives) – type projects.

Figure 3. PFI Model Structure (Akintoye and Chinyio, 2001)


It’s important to note here that the bank is the senior debt provider. This is generally the case but not always. This is an important point to remember with these projects, which are often not as liquid as smaller investments which might be part-funded by banks. Collateral in the form of a bridge or road project, for example, might not be attractive to banks which have provided funding for the project in question.

The nature of Italian Public-Private Partnerships (small population of projects relative to other countries, infrequent use of capital markets for financing, lack of transparency in the tender process) means that the task of analysing how debt is issued by banks is made more difficult. Nonetheless, this paper will put forward several justifiable theories as to what factors underlie the process. In turn, the factors to be analysed are: risk, firm size, the length of concession period and the interest rates on date of financial close.

Firm Size as a Factor in Banks’ Willingness to Issue Debt for PPP
In a sample of 75 Public-Private Partnerships (see appendix 1), what is notable how large the firms’ are in terms of market size. The reasons for this could be manifold, including existing relationships between banks and these firms, a long track record of large infrastructure projects having been delivered by these firms, the criteria of the project demanding larger private partners or just a lack of smaller firms competing for the tenders of the Public-Private Partnership projects in the sample.

In the 75 projects sample mentioned above, the average firm was several billion euro in terms of market capital (where some firms were the main private partner on more than one project). This would comfortably put these firms in the top decile of firms in the European Union and would suggest that firm size is a highly important factor when banks are deciding whether to issue debt for the project. Particularly given that some of the project sizes could feasibly be taken on by much smaller firms.

Regarding how smaller firms could take on many of the projects in the sample; this is confirmed by running a correlation between the main partner firm size and the debt required for each project. There is virtually no correlation at all between the two.This would seem to give further weight to the theory that banking institutions prefer when the main partner is a large company.

The sample of 75 projects is almost entirely populated by projects where the main partner is an Italian firm. There are some exceptions, however. These are provided by Barclays Bank, the Sunpower Corporation and the Foresight Group. While the Foresight Group is estimated to have a shareholder equity value of €60m, Barclays and Sunpower Corporation are among the largest firms on the list, suggesting that when banks do look to foreign partners, the size of those foreign partners is a major consideration.

Likewise, Akintoye et al (2001) note that “the credit risk associated with the borrower is of little importance and the finance must be judged almost entirely on the basis of the risks that may threaten the project completion and operation.” This would suggest that, where funding banks are concerned, a track record is highly important. Larger firms, all things being equal, will tend to have longer track records than smaller firms, thus providing further inclination for banks to provide debt to them over smaller firms.

As a final consideration, it is more difficult to ascertain the importance of this particular measure (firm size) in banks’ decisions to fund the Public-Private Partnership without access to competing bids (i.e. those that bid but did not win the tender). The true measure of the importance of firm size in banks’ decisions to provide debt or not would require the full list of applicant firms for each project, along with their estimated firm size.

Length of Concession Period as a Factor in Banks’ Willingness to Issue Debt for PPP
Although the length of the concession period is technically part of the risk profile of each project, there is more at play from a banks’ perspective than just risk when it is being considered. There are potential agency problems, issues about capital ratios and considerations to be made on how they view the long term interest rate curve. In fact, the interest rate curve has to be a major decision for banks in looking at the length of the concession period for each project.

The interest rate curve can tell banks a lot about which medium- to longer-term (as projects in Public-Private Partnerships tend to be). The interest rate curve at any particular time should offer a good indicator of whether banks will offer for longer-term projects or not. Typically, they would face an upward facing interest rate curve (that is, interest rates in the longer term are normal) whereas at times – such as during the recent global financial crisis – they will face a flat interest curve, indicating ongoing uncertainty in the financial industry and probably some reluctance to become involved in projects with longer concession periods.

From a banking perspective, the disadvantage of many Public-Private Partnership projects it that the length of the concession period can potentially create agency problems that might not otherwise appear in shorter-term projects. These are the same agency problems that are created by mortgages; bank management often do not have the time in office that it takes for the projects to be realized, so there is some misalignment between the duration of management and that of the debt being issued.

The concession period of a bank’s existing stock of loans should also be a factor in their decision to provide debt or not. Banks don’t want to be over-loaded with long-term loans, given potential liquidity problems that can be created in such a scenario. Therefore, as a general rule, they attempt to balance the concession periods of their loans. On this basis, every time a project is being analysed, it is also being analysed (or at least, should be) in conjunction with existing loans on the bank’s balance sheet.

Interest Rates as a Factor of Banks’ Willingness to Issue Debt for PPP
Interest Rates in the European Union have been set by the European Central Bank since the mid-1990s. As Figure 2 above illustrates, these have been at historically low levels for almost five years, meaning banks have had access to increased liquidity and perhaps some incentive to provide value-creating loans during the same period. In fact, when one considers how the Public Private Partnerships projects held up during the past number of years next to the economy, this would seem to hold true.

When banks are considering the interest rates as a factor to accept to fund a project or not, other projects’ interest rates are also a factor that should be considered. When two projects are deemed similar but one has a higher interest rate than the other, it doesn’t necessarily mean that the higher interest rate project will be funded over its competitor. The bank will also figure how likely the project is to maintain payments at higher interest rates over the concession period.

The interbank rate is a fundamental factor in how banks made lending decisions. As Figure 4 which follows illustrates, since the onset of Public-Private Partnerships in Italy, 3-month interbank exchange rates in Italy have been on a fairly consistent downward trajectory. This suggests several elements are at play but primarily that Italian banks have more access to liquidity than they did in the time before PPP (and thus can take on illiquid projects such as those of project finance).

Furthermore, should the interbank interest rate be relevant to the issuance of debt for project finance, one would expect the number of PPP deals to fall to some extent as the rate rises. That is to say, the two have a negative relationship: one tends to rise as the other falls and vice versa. This is broadly consistent with what seems to be happening when Figure 4 is compared with Figure 1 from previously. In particular, it is notable how PPP plateaus in 2009 as interbank rates rise.

Figure 4. Historical Italian 3-month Interbank RatesSource: Banco d’Italia

Inflation rates also play a factor. These are accounted for by the real interest rates in Figure 2, but nevertheless, a graph is indicative. Figure 5 below illustrates historical Italian inflation rates from 1995 to 2013. Most Central Banks target an inflation rate in the region of 2% annually and Italy has been very close to the target since PPP projects began in 1994. Given that Italy’s average inflation from 1958 to 2014 was 6.58%[2], it could well be that the low-inflation, low-interest environment has combined to make a set of circumstances which are amenable to project financing. This is what economic logic would lead us to believe.

Figure 5. Historical Italian Inflation Rates.

Source: Banco d’Italia

Solidity as a Factor in Banks’ Willingness to Issue Debt for PPP
The solidity of the main partner – which is the current equity position of the main partner – can be used as a proxy for the risk position of the firm. It follows that firms with more solidity should be willing to take on further debt. Or at least, they should be in a position to take on more debt if the occasion arises; whether or not management of the main partner are willing to take on the debt, is a matter of discretion.

Nowadays banks have proprietary risk-measurement tools (as permitted by the Basel III Accord), which means that each banks attributes different weights to different factors when measuring risk. Given that the weights behind these proprietary models are not disclosed, it is difficult to say too much about banks’ appetite for risk. It is suffice to say that they wish to minimize the risk. Maximizing their equity – four the purposes of this paper, the solidity – is one way of doing so.

Grimsey and Lewis (2002) note that PPP projects are characterized by all downside risks for banks, with very little to compensate on the upside: “the facilities often do not have a capital worth, in terms of a wide market, to which lenders would attribute value.” The authors seem to be suggesting that PPP is characterized by large liquidity risks, which would be another reason to opt for larger (more solid) SPVs. This view of risks taken by the banks is surely another reason for them to be reluctant to hold equity in the projects.

Akintoye et al (2001), note, “financial companies are involved in a rigorous process of project’s risks evaluation and subsequently they refine the terms and conditions  of their commitment…During this process, the finance companies utilize a range of tools to ensure effective fund provision, contract enforcement, sharing their market information and risk management skills.” In light of this, it would seem that banks are highly aware of the risks inherent in providing funding to PPP projects.

As the number of Public-Private Partnerships increases in Italy and as they become a mainstay on the Italian banking industry’s funding horizon, experience would suggest that banks will expose themselves to further risk with these projects. There is a marked tendency to take on more risk where one hasn’t been burned before.

Conclusions
As the recent global financial crisis illustrated in stark terms, banks are not immune to making irrational funding decisions. Given how low interest rates are currently set, many banks (although not all), for the moment at least, can make loans they wouldn’t otherwise make should interest rates be at traditional more higher levels. The potential for banks to make irrational funding decisions correspondingly becomes easier in such an environment where money is cheaper.

Italy didn’t escape the global financial crisis but did become one of the notorious PIGS[3] group, which were dogged by high public debts. These high public debts in Italy might also have contributed to a slow-down in projected infrastructure development by the government and on the private side, a reluctance to invest in projects where the partner was – by some predictions, at least – due to go bankrupt in the near future. In this environment, reading too much into Public-Private Partnerships in Italy can be unwise.

Nevertheless, the advent of Public-Private Partnerships in Italy in the 1990s offered a new avenue for Italian public projects on one side and an avenue to diversify funding opportunities for Italian financial institutions on the other. On the basis of financial theory, the initiative made the funding of Italian public projects more efficient. Likewise, judging by the growth in the industry in the intervening period, it is fair to say that the initiative has been well-received so far.

This study has put forward some theories about the decisions banks take while funding those same projects. Without data into the projects that don’t become funded and greater transparency in general on Italy’s Public-Private Partnerships, the research remains largely theoretical, rather than empirical. However, the small data sample did provide some insight into where funding decisions are being made by banks for Italian PPP. These data seem to suggest the following:

-          Italian firms are preferred as the main partner for funding.
-          Larger firms are preferred for funding.
-          There is no virtually no correlation at all between major partner size and project size.

Bibliography
Akintoye, A., Beck, M., Hardcastle, C., Chinyio, E., Asenova, D., (2001a). “The Financial Structure of Private Finance Initiatives.” 17th Annual ARCOM Conference, 5-7 September 2001, University of Salford. Association of Researchers in Construction Management, Vol. 1,
361-9.

Akintoye, A., Beck, M., Hardcastle, C., Chinyio, E., Asenova, D., (2001b). “Risk Identification Practises under PFI Envionment.” 17th Annual ARCOM Conference, 5-7 September 2001, University of Salford. Association of Researchers in Construction Management, Vol. 1, pp.875-883.

Akintoye, A., Beck, M., Hardcastle, C., Chinyio, E., Asenova, D., (2001c). “Management of Risks within the PFI environment.” 17th Annual ARCOM Conference, 5-7 September 2001, University of Salford. Association of Researchers in Construction Management, Vol. 1,
pp. 261-70.

Broadbent, J., Laughlin, R., (2003). “Evaluating the Private Finance Initiative in the National Health Service in the UK.” Accounting, Auditing and Accountability Journal, Vol. 16, No. 3, 2003, pp. 422-445.

Esty, B. C.(2003) "The economic motivations for using project finance."Harvard Business School 28.

Grimsey, D., Lewis, M., (2002). “Evaluating the risks of public private partnerships for infrastructure projects.”  International Journal of Project Management 20, pp107-118.


[1] http://www.economist.com/blogs/schumpeter/2011/07/italys-finances-0?fsrc=scn/fb/wl/bl/pubskittlestheitalianversion
[2] http://www.tradingeconomics.com/italy/inflation-cpi
[3] Denominating Portugal, Italy, Greece and Spain.

No comments:

Post a Comment