Chris Vermonts interview for the 'Infrastructure Journal' publication.
Burgeoning competition from Development Finance Institutions, risk aversion of hard-currency lenders and the limited appetite of local banks makes project financing in infrastructure markets of emerging economies ever more challenging. Infrastructure Journal's Jérôme Clarisse met Chris Vermont, Head of Debt Capital Markets at Frontier Markets Fund Managers, to get his thoughts on guarantee provision for nascent infrastructure projects located in some of the world's poorest countries.
A seasoned professional with over 20 years experience of structuring and syndicating finance for projects in emerging markets, Chris Vermont joined FMFM in 2007. He is responsible for GuarantCo, a US$400 million fund which is sponsored and set-up by the governments of Netherlands, Sweden, Switzerland and UK, to provide local currency guarantees to enable finance of infrastructure projects in low and lower middle income countries. FMFM acts as adviser to GuarantCo and Vermont's role includes new business origination, capital markets products and leveraging GuarantCo's capital structure. He was formerly Head of Project and Structured Finance and latterly Head of Debt Solutions & Distribution at Australia & New Zealand Bank.
Jérôme Clarisse: Could you tell us more about GuarantCo and its inception?
Chris Vermont: GuarantCo is among a number of initiatives under the umbrella of the Private Infrastructure Development Group (PIDG), which is an association of seven European governments, the World Bank as well as a number of organisations with observer status. The group includes several private infrastructure initiatives, focussing largely on Africa though a few - such as GuarantCo - have a broader remit. The rationale behind these is to provide infrastructure finance outside of a normal DFI framework.
Each of the PIDG facilities sponsored by the governments is managed by a private sector manager using a mix of private and public sector funds. The idea behind it is to inject greater commerciality. Earlier, funds were quite straightforward. For example, the Emerging Africa Infrastructure Fund provided US dollar or Euro denominated loans to African projects. However, a few years later, PIDG expressed a desire to address local currency finance and created GuarantCo. However, rather than establishing us as a local-currency lender, which in theory they could have done by fund-raising locally, PIDG recognised the risk of crowding out the private sector. Since the ultimate objective was to encourage local capacity building, GuarantCo was set-up as a guarantee organisation. It means that on every transaction, we have to work with a funding source, be it a local bank or an international bank operating locally.
JC: What exactly GuarantCo's involvement within the financing of infrastructure projects?
CV: Broadly speaking, we are a single-purpose entity offering guarantee products. Occasionally, this might be in the form of insurance, even though we come from a banking background which is used to the provision of guarantees. Beyond that, we have a fairly wide interpretation of the sectors in which we operate. Although we talk about infrastructure, this might include the components of infrastructure such as cement or steel, or other fabricated items. Furthermore, the type of support can vary, whether it is guarantees for loans, securitisation, mezzanine finance or senior debt. The key underlying feature is that it has to be helping infrastructure in low income emerging economies.
JC: GuarantCo's target list of countries typically extends to most emerging markets in Latin America, Africa and Asia. Yet it excludes countries where the PPP model is tending towards maturity and has been successfully adopted. The examples of Mexico, Brazil or South Africa spring to mind. Is this a deliberate choice?
CV: It is a function of where our capital comes from. We have four European government shareholders and the funds come from these countries' aid budgets. Their ambition is to see the poorest countries served. It creates some dynamic tension because, as you rightly say, the markets are not necessarily as well developed and we have to be realistic about what we can achieve. However, we have operated in a number of unsophisticated countries in terms of capital markets, for example - Chad, Palestinian territories, Uganda and Cambodia. These are countries where it is possible to do business even if the bank and capital markets are relatively primitive.
JC: Of the countries mentioned, do you view any specific markets as most promising going forward? For example, I note that Egypt falls within your radar.
CV: Well the problem with Egypt is that so far, we have not been able to add very much value as the local banks are pretty enthusiastic already. We looked at Cairo Waste Water [Project Database] for example but banks were already doing 15+ years' financing at very tight margins. Where we do not add value, we do not seek to spend much time. It might be that other projects in Egypt which have less of a state wrap around them, will be more appropriate for us. You could argue that India is also a very developed market. Therefore, we will not play in standard project finance there. In the Indian market, it would have to be in a niche where for some reason, absence of support means that GuarantCo could add value. So among projects where we have played a role, I would mention the Calcom cement plant in Assam - a state not covered by the main India project lenders, or a slum rehabilitation project in Mumbai. Most banks do not see such projects as central to their financing agenda. In Thailand, we are looking at a specialist steel pipe project. It is difficult to do steel projects in Thailand due to fresh memories of the Asian crisis and all the money lost by Thai banks in the steel sector. But I think that the honest answer is that there are not that many countries where we operate, which have that much of a developed PPP-type of approach. They have not yet followed the British or European model.
JC: Is there a clear and visible trend in this direction though, in any one or two of those countries?
CV: Yes certainly. I believe natural budgetary constraints indicate trends will go in this direction. It is most apparent in the power sector. But unfortunately for GuarantCo and due to historical reasons in Africa and elsewhere, there has been a tendency for power purchase agreements to be signed with dollar indexation, which is quite unhealthy. When you have a savage devaluation that has happened in Pakistan or in Indonesia during the 1990s, you can bankrupt a project very quickly. Even if the offtaker is contractually due to pay a higher tariff owing to dollar indexation, the contract is often simply abrogated as a result of unaffordable payments.
In Africa, it is a great disappointment to us that there are still so many projects which get done in Euros or US Dollars when ultimately, revenues are denominated in local currency. It is difficult to get people to change their ways. Local utilities have been told for years that the only way they can get finance is by indexing their tariffs or PPAs to hard currencies. Our response is that the case for local currency is nonetheless valid, even if only as a part of the off-take agreement. There has been a slight change in this direction but the problem is that the main financiers behind those projects are traditionally either DFIs or international commercial banks, which typically want to lend in Dollars or Euros. It is only in the very large markets such as China or India where one could find a real preponderance of local currency, because these markets are deep enough to provide it. Our goal is to use our guarantees to improve local currency terms, especially tenor, so that local currency is a viable alternative.
JC: As a guarantor of local currency debt issuance, which obstacles do you identify as being particularly prominent in your focus markets - be it from a regulatory, legal or perhaps even cultural standpoint?
CV: To begin with, I would say that we should not over-emphasise the difficulties. More often than not, reasons why banks do not go to Africa - or certain of the smaller Asian countries - are not actually risk-related but rather have to do with the fact that those markets are just not significant enough. On the basis that most banks do not want to do a project which has a capital expenditure of less than US$100 million and that in certain African countries projects tend to be on the smaller side, you might only ever have the prospect of less than 2 or 3 projects a year actually making it to financial close! It should then come as no surprise that commercial lenders would only really want to be in markets with large volumes of decent-size transactions, and therefore shun our markets.
There are many countries where the risk profile is not so bad. For example, throughout West Africa several countries follow a similar Civil Code to the French legal system. However, inevitably you do get certain countries where the regulatory regime is hostile to our work. For instance, China - which ironically we can do because its per-capita income falls within the remit of GuarantCo's target country list - actually has specific laws preventing guarantees being issued cross-border, unless you have a local presence. Similarly in other countries, central banks may be extremely conservative and may not recognise the full value of the guarantees and therefore give the appropriate sort of capital relief. As you mentioned, the general legal system may sometimes just not be sufficiently well-developed for limited recourse project-financed structures, and for the possibility of securitisation. Even where laws exist, the implementation is can be poor.
JC: Would you say there is sufficient appetite from local funders towards local currency lending in project-financed transactions?
CV: This is another limitation. However, relatively speaking, it has not been as bad as it used to be in the light of the generalised funding problem more recently. But yes, it has been historically problematic particularly in Africa, to get banks and financial institutions to lend sufficient tenor for infrastructure projects. Try financing a power project over five years and it is quite difficult to make the cashflows work! So you really need 10+ years for these types of projects, and very often the local banks are not used to this due to their asset-liability mismatch. The way GuarantCo caters for this is to have guarantees that not only cover project risk but also funding risk. So if the bank cannot refinance itself due to a funding crisis, then it can call on the guarantee in specific circumstances.
JC: Coming on to your process of investment selection and decision-making: firstly what is the quality of leads coming to you? Secondly, to what extent are they viable and bankable investment opportunities? Also can you tell us about your credit committee approval process?
CV: As I am sure any fund manager will tell you, the number of projects which get done versus the number of projects which get pitched to us is a very small proportion - probably 1 in 25 as an approximation. Even those projects where we get approval, quite often fall by the wayside before they reach financial close. GuarantCo's approval process is relatively streamlined compared to a bank. We have a two-stage approval: firstly we have a New Business Committee which gives an in-principle approval and points out any issues for due diligence. Further to this, we have a full credit paper which is prepared and is approved (or not) at that stage. We do not require separate approval for asset writing or country risk for example, which many banks would normally have. I should point out that credit approvals are formally taken by an external Board of six non-executive directors plus a representative from our leverage providers. So we as fund managers do the work, prepare the papers and make recommendations, but final decisions are left to an external board composed of directors based in various parts of the world. The Chairman is ex-IFC; the second most experienced director is chief executive of a private sector Middle Eastern bank; another director is a fund manager based in Shenzhen; yet another one in the UK. The Board's role is to represent the interests of our shareholder governments and to balance development and commercial goals.
JC: Earlier you mentioned the Calcom cement plant in Assam, India. Could you give further details of this and transactions in which the fund has participated lately?
CV: The Calcom plant is a Greenfield project for the build of a cement plant in Assam. Cement is usually not a problem in most states in India but if you are a relatively small to mid-sized producer located in one of the more difficult states like Assam, bank finance is unavailable since it is rationed out to the major business houses only. Quite frankly, the banks do not have much interest in going all the way up to the north-east (of India). The bank that had been syndicating this particular US$120 million transaction had a failed syndication, and the financing could not be completed. They came to us and we put in a guarantee for just over a third of the debt, which then meant that there were enough banks which could increase their participation to reach financial close. This was a situation where although there was German equipment going in, the ECAs were not comfortable with Assam or with the developer's size so that the local banking market provided the only financing solution.
A second transaction was a telecom project in the Palestinian territories, which closed about a year ago. Given the limitations on human movements in Palestine due to Israeli road blocks, mobile telephony is incredibly important. The problem is that the incumbent operator has faced difficulties modernising its network. So this second operator, Wataniya Mobile [Projects Database] was put forward and promoted by Q-Tel from Qatar, and with local investment in joint venture. The entrant pumped in a lot of money - $200 million or more - but also wanted to have a debt element. However, the lack of commercial banks in Palestine made this difficult. So there was a combined Ericsson and Standard Bank credit of $22 million against EKN's (Swedish Export Credits Guarantee Board) guarantee. We then had a $33 million domestic facility which was partially backed by GuarantCo, and then a $30million IFC loan. The sponsors very much wanted local Palestinian banks to be involved, with a view to increasing Palestinian involvement for obvious political reasons.
However, local banks were not into project finance. In this case EKN could only guarantee offshore banks, not local ones. So GuarantCo was the only guarantor around for this. We also had a fairly interesting financial close, as bombs were going off in Gaza in February 2009. Fortunately the project was in the West Bank but we had lawyers who were in Gaza, literally in bunkers and saying to us: "Sorry we have not yet responded to the documents with our questions, due to lack of internet access. We could go across the street to the UN but we might get shot at!"
JC: We discussed previously a number of reasons why global commercial players may not want to get involved at the frontier end of emerging markets - lack of deal flow, limited ticket size or domestic currency denomination. How much of your work is actually done with larger commercial players or equally, with Export Credit Agencies?
CV: We do work with ECAs but not as much as I would like. However, they do all have local currency programmes, so it is quite possible to cooperate. In the last 3 to 4 months, what has been interesting is that a number of banks have now started to form DFI and sovereign-type coverage groups. Certain banks like Citi have always had a DFI coverage function but just in the last three months, Standard Bank has announced the reorganisation of its DFI coverage, Deutsche Bank has formed a DFI coverage team and BNP Paribas is currently in the process of doing so too. Because the syndicated loan market is still relatively inaccessible for emerging markets, players who are slightly ahead of the game have worked out where the capacity exists to actually place deals. I think they are all coming to the same conclusion, though quite separately and by following different business models.
JC: As a general comment, the evolution of pricing spreads in project finance markets has been upwards with the unravelling of the global crisis. Looking at your markets, would you confirm this view?
CV: Pricing has definitely edged up in our markets, although again it was already relatively high so that there has not been as stark an increase. What has been particularly interesting is that we had always assumed that pricing in Asia was going to be lower. Taking the discussion into context, when I worked in commercial markets, you would typically see lower pricing in Asia than you would in Africa. But this has now been reversed. Indeed, we are actually getting more favourable spreads in Asia than we are in Africa at the moment. This maybe down to the fact that Africa is still relatively dominated by the DFI community and the DFIs could perhaps be accused of under-pricing sometimes. There is quite a lot of competition among DFIs to try and find good projects in Africa, since the volume of projects making it to financial close is far from substantial. So whereas we might be seeing projects paying guarantee fees of 300-400 basis points in Africa, most of our Asian projects have been closing at 450-550 bps levels for risks which are in fact lower. That said, I think pricing is beyond its peak.
JC: Finally, the recent crisis has seen deleveraging of debt-to-equity ratios used in project financing transactions, owing both to risk aversion and scarcer liquidity. How does this compare with your own observation of GuarantCo's focus markets?
CV: This is true. In any conversation with a sponsor, it is currently more acceptable to demand an increase in equity when a project is perceived to be under-capitalised. That said, I do not think that projects at the frontier end of emerging markets were very highly levered in the first place. It has not been uncommon for us to see 60:40 or even 50:50 types of structures being used over the last 5 years, and this has not changed significantly. In any case, I do not think we have worked on a project which had more than 70:30. You would not really expect to see those levels in our markets whereas in developed markets or certainly in UK PFI, leverage has traditionally been high. However I'm not saying that high leverage necessarily leads to poor projects. But our motto is definitely "good projects in poor countries rather than poor projects in good countries".