We conducted an interview jointly with Loïc De Cannière, Managing Director of Incofin, the prominent microfinance investor and fund management company, and Lieven Dupon, Senior Underwriter at ONDD, Belgium’s national export credit and political risk insurer. The result is a candid discussion, interesting and informative, and not just with respect to coverage for microfinance investments.
Mac Johnston (robert wray pllc): Incofin has a global PRI policy with ONDD covering microfinance exposures in a number of countries. Was this a first for both of you – the first PRI policy for Incofin and the first major policy covering investments in microfinance institutions (MFIs) for ONDD?
Loïc De Cannière (Incofin): Yes. I knew political risk insurance from my previous job, and when I started at Incofin, I explained the advantages of a PRI policy, and this was absolutely endorsed by the board, mainly, I think, for the aspect of transfer risks.
Lieven Dupon (ONDD): Of course, as an ECA, we have been offering PRI cover for a long time. Incofin was our first MFI client for PRI policies. In the first phase, Incofin made the decision on whether or not to be covered on a certain transaction, on a case by case basis. The big boost on the relationship between Incofin and ONDD was the introduction of the concept of having a portfolio cover whereby, you can state that any transaction that fits the policy criteria benefits from the policy, and this also was attractive as to pricing.
Mac: Other than transfer risk, what were Incofin’s concerns about political risk that they wanted to see covered in the policy?
Loïc: We thought that the risk was high in certain countries; however, after a while we understood that we are not specialized in assessing the political risk of a country as we are in assessing the financial risk profile of a microfinance institution. So we were very happy that we were able to convert the coverage into a portfolio insurance. Of course, we continue to assess the political risk of a country – but let’s say that we feel much more comfortable with this. And after all these years, I think we are satisfied with this because we are investing in difficult countries, and in many of those countries, incidents took place while we were there. Take the [Russian] invasion in Georgia; the uproar in the Rift Valley in Kenya; and, of course, Nicaragua. At a certain point there were problems in Guinea with the transfer of funds.
Mac: What is the tenor of Incofin’s policy?
Lieven: The concept is really a portfolio cover, which is to be reconfirmed each year, but it has the system, which I think in English they call it “risks attaching,” so in a certain policy period, for example this year, you attach risks on it with a certain tenor, for example five year transactions, or three years or whatever, and these are covered. And if we were not to [renew the policy], for example, next year…we cannot withdraw on that five year transaction.
Mac: How about equity?
Lieven: For equity transactions on countries open for ONDD cover we can offer long tenors. I mean, I think we have never had an issue with the fact that Incofin is saying “I want cover for ten years,” and we are saying we cannot do it.
Loïc: No, but in some countries you put restrictions on tenor of the insurance. Like when we want to extend a three year loan in Tajikistan you would say “No, we would only offer you cover for one year,” and then we get into problems. So we need to structure a loan which is callable after one year, that kind of thing.
Lieven: And that’s because if you have countries whereby there are transfer restrictions and we are off-cover for medium-term deals, the solution might be that Incofin offers one year deals or 18-month deals and then we can follow. But that is indeed because of the country risk appetite, not because of statutory restrictions.
Mac: Was the original ONDD policy “off the shelf” or was it tailored to MFI risks?
Lieven: Yes indeed, it was a classic, standard wording. But I have to add that we are currently in the process of adapting the insurance cover to the microfinance industry and this is a process that has been going on for a couple of months and now we’re in the final stage of this. We have learned from last year that the policy needs some adaptation to the microfinance industry, in order to clarify whether or not certain events can be interpreted as, for example, a government action event. I think that it’s a natural evolution that you start with a client or with an industry covering it with your standard wording and when the scope becomes larger and larger and you learn from experiences, that you try to bring some tailor-made elements into it. That’s basically the message I want to bring here.
Mac: Ok. At whose initiative was it – at yours or Incofin’s?
Loïc: I think it came as a consequence of our discussions on a specific case, where we discussed whether there was a link between some political event and the defaults of a microfinance institution and whether – yes or no – you could say that the default in the microfinance institution was a consequence of political events. In the course of that discussion, we became aware, as the insured party, that political events were considered in the policy as physical destructions of agencies or branches of a microfinance institution and with violence and everything, whereas situations can be much more subtle and much more sophisticated. The political behavior in a country can lead people, for several reasons, not to repay and progressively run the institution into bankruptcy and this is of course the difficult area which needed to be clarified among us – and which also caused some, well, some discussions and some tensions also between us – whether or not the political risk insurance was applicable.
Lieven: And to add on that, I think it was also a kind of a wake-up call whether the insured and the insurer both understand the same thing. I mean, do we understand each other on what is covered and what is not covered, and that leads to more detailed discussions. You don’t have a miracle solution, of course. Some cases will be straightforward, like, for example, the non-transfer cover: either you can get the money out of the country or you can’t. But especially as to government action/inaction, things like that, we will try to specify things. It will always remain a difficult area, whereby you will have discussions and also, in a way, base yourselves on trust that one party or another will act in a reasonable way in assessing it, because these are not black and white situations.
Mac: It is known that ONDD paid a claim under the Incofin policy for a loss in Nicaragua. What can you tell us about that experience? Did you learn something important in the process?
Lieven: Well, what can we disclose on that? What did we learn? We learned a number of things. We learned, for example, that one important risk in the industry is the phenomenon of the non-payment movement, which we weren’t aware of or we didn’t feel it was a major risk. That was new to us. And linked to that, the fundamental question: to what extent can you call a non-payment movement a political event, because you have a non-payment movement in Nicaragua but in other countries you have similar types of pressure groups.
Loïc: One major one also in India…
Lieven: In India, yes. So you can question yourself: is this a political movement or would you rather call it a social movement or whatever? So there you come to the real fundamental question: should this be regarded as a political risk? That was the first challenge, I think. That was a very important challenge. And I think another element is the same as interest rate caps, which was also for us a kind of a wake-up call: whoa, this is a phenomenon happening not only in Nicaragua, but in some other countries as well. And with the same question, should the policy pay – yes or no? And then you come to the question: is an interest rate cap reasonable or not? Because any country, including Belgium, imposes interest rate caps on, for example, consumer credits, so there you come to an interesting question: to which extent should the insurance policy pay if a government – even a democratically voted government – imposes democratic measures through parliament? So this case really made us think on how to address all this.
Loïc: Yeah, I think that what was important in Nicaragua was that at a certain point of time you acknowledged as ONDD that the movement had a clear political dimension because it got the explicit support from the president, a law was debated in the parliament, and now it has been approved – so this was, as I understand, Lieven, for you, the critical point to say yes, it’s not only a social movement. It’s a political movement.
Lieven: Indeed. And I think there it’s interesting for clarifying the policy for the future that our view is that a political risk insurer should not cover a social movement, for example of farmers in some region opposing on paying some of their loans because they feel it is too expensive. This is not political. But if of course the whole country is involved, including ex-presidents and presidents and politicians et cetera et cetera, to a certain extent you have to judge OK, is this becoming a political matter? But that is not an easy exercise.
Loïc: An interesting case is now something which happened in India – without any consequence on our portfolio, Lieven, so don’t be preoccupied [laughs] – where the religious leaders in Karnataka were calling not to repay the loans. There’s also a “no-pago” movement in India that’s mixed with religious elements, so this is a very difficult situation to judge: is this political or not? Ok, there have been no statements made by politicians as such, it is a mix of religion and no-pago, but of course we would like to see this covered by an insurance, but I understand that it will not be easy.
Lieven: No, that will not be easy because with the Nicaragua case, such non-payment movements reflecting anger or mistrust of the industry by borrowers is a risk inherent to the industry, but that cannot be regarded as a political risk. Except if indeed you have political driving forces behind it, which is difficult to prove. In the Nicaragua case we came to the conclusion that these impulses from the political world were clearly there. But as a starting point, a non-payment movement is something that is not, should not be covered by a political risk insurer. That’s really the lesson we learned. And another thing in Nicaragua was that you had political violence, which is a classic cover element, but was this political violence an important element, or was this a kind of side phenomenon? Was the major issue that people simply do not want to pay also motivated by the fact that there were rumors on interest rate caps? Or even debt abandonment imposed by government? Or was it more the violence that created the crisis? These are difficult questions and in the end, I think you need to come to a kind of a judgment, which in some cases is not black and white, but you have to come to a kind of agreement with the insured that this is not black and white, this is grey. Let’s adopt this position and then try to move on. That’s basically what we did. And I think in that way we, towards our insureds, try to act in a reasonable way and I think that the trust factor is also important because these are not easy insurance policies to understand. It’s not like fire insurance you buy on a house. So you have to understand each other; you have to trust each other in this kind of business, I think.
Mac: That’s an excellent point, and as you know, it’s not a relationship that’s universal between the parties and it speaks well for both institutions. Lieven, you have indicated that ONDD would consider offering coverage against the imposition of interest rate caps. How about that and also how about political violence coverage that affects the ability of a substantial number of downstream borrowers to pay the MFI, so as to impair its viability. That’s a tough one, isn’t it?
Lieven: We are in the process of considering cover for interest rate caps, as we feel that it is a political risk. It is linked to governments, with one important nuance of course, that it should pass the test of being unreasonable and that’s the difficult part, of course. And there you have to turn to the classic PRI exclusion language that stating that non-discriminatory/non-arbitrary actions that a government normally takes in the public interest, should not be covered by the policy. In other words, if a government – and I think Ecuador has done it – if they impose interest rate caps of 30%, I mean, the MFI sector has to live with that, just as the Belgian government is here imposing caps on consumer credits. And of course there, you come to grey zone areas: is 20% reasonable or unreasonable? There you will have easy cases, whereby for example if a populist dictator all of a sudden decides that microfinance loans should be for free from now on – that’s an easy case; this is an unreasonable action. Imposing a cap of 20%? Yeah, you can question that. So you will have grey zones, but I think by stating interest rate caps are in the area of political interference, but with that nuance…that’s the best you can do, I think, in the insurance policy. And then for political violence…
Loïc: Yeah, that’s an important one!
Lieven: Well, Loïc, we will cover that as well, but a normal political risk insurer will say, as to political violence: if you don’t have damage to the MFI, I mean damage to the assets, basically, it’s not covered. You need direct damage due to political violence. Here of course we need to move to situations whereby you don’t have any damage to the MFI branches, for example, but you have a situation of war or civil war whereby borrowers, simply because of that situation, are unable to pay the MFI and the MFIs are going bust. So this is also something we want to include into the policy text to avoid discussion afterwards. But again, we will not be able to avoid all grey zone areas and there are no miracle solutions. What we will try to do is include that concept into the policy.
Loïc: Parallel to the interest rate caps is a situation – I’m sorry to come up with all these examples, but this is real life – where the government decides to create a state-owned bank, a microfinance bank, and starts extending microloans at an interest rate of 1% per annum, financed with cheap money, thereby actually destroying the whole microfinance market and the whole repayment culture. Is that something that would be covered under the new policy? It’s like interest rate caps but it’s another way of intervening in the markets.
Lieven: I must admit that I haven’t given it a decent thought.
Loïc: It is without consequences so far, without substantial consequences, but actually the model itself is, in my opinion, extremely dangerous.
Lieven: Yeah. And is it a concrete threat?
Loïc: Well, it’s materialized. I mean, the bank exists and they started extending loans with money being financed, but it’s without substantial impact on the market so far. It has created some concern, yes.
Lieven: It’s an interesting question and I’m not able to give a response on that. But I very much understand that this is a real, real concern for anybody that is in this industry. Because if they have the capacity to rule out all the normal players…yeah, that is a big question mark.
Mac: OK, thank you. A typical MFI might go through stages, progressing from an unregulated entitiy to a deposit-taking, regulated financial institution. How should PRI deal with regulatory risks along the way?
Loïc: I think that regulation is absolutely key in all stages of the lifecycle of an MFI and any unreasonable behavior – of course, if it is discriminatory behavior, it’s absolutely inacceptable – but even if it is not discriminatory, but if it is unreasonable, it’s difficult to live with, and this is something that we consider as a very high danger. Or the lack of regulations, or of intelligent regulation. So I would consider this as a very high risk together with everything which is linked to no-pago, non-payment movements, dependent on whether you qualify them as social or as political movements.
Mac: It’s a difficult area because insurers typically don’t want to say that we’re going to make the situation better than it is. In many cases you’re moving into an area where there really is nothing.
Loïc: Exactly. I understand the point.
Lieven: Maybe it’s a question for robert wray, because I understand that your law firm is involved in providing a lot of legal advice on, for example, MFIs that move from unregulated lenders to regulated lenders.
Mac: What else would Incofin like to see political risk insurers offer in terms of coverage for investments in MFIs that you presently can’t obtain?
Loïc: There’s one thing – and I know this is not covered by the PRI policy which we have with ONDD – which is acts of God, which under other types of policies are covered but not under this policy and I never understood why. We have invested in Haiti and our investment has been destroyed, literally – I mean, the buildings have been destroyed and everything. Luckily we have another insurance cover with the Flemish government which included this act of God, but I would be very much in favor of acts of God being included in the PRI policy. We were close to having invested in Pakistan and I’m happy now that we didn’t – what would have happened if we had done that and the institution would have been flooded? Would that have been a part of the PRI policy? These are essential points which need to be clarified in order for it to be a really convenient and adapted policy.
Mac: That’s an interesting comment. I think that most political risk insurers will probably say, “That’s not the business we’re in.” On the other hand, there really are some similarities between natural disaster insurance and political risk insurance, in the sense that they both cover catastrophic events.
Loïc: [jokingly] You think he got the message? Lieven heard the message, I hope.
Lieven: I think acts of God were actually covered 10 or 15 years ago in our political risk insurance cover, but we removed it because we have seen that it is indeed more a cover that is offered by property insurance markets and no PRI insurer is offering it. It is really something for the property insurance market. That’s why we left it out. But indeed, I note now that Loïc, you have two very interesting examples of Haiti and Pakistan. I can imagine that for you it of course would be very practical to have it included just in one policy instead of trying to cover it in some other insurance market because that would render the whole thing much more complex for you.
Mac: Lieven, has ONDD written MFI policies for other institutions, and are you particularly interested in doing so?
Lieven: Yes. I think it is mainly also thanks to Incofin because Incofin is also acting as a fund manager in a number of funds We have been providing insurance policies for various other funds and institutions. In Belgium there are quite a lot of institutions involved in the industry – not necessarily for huge amounts, but yeah, you have a number of financial institutions, other institutions, NGOs, that have activity in that industry and which we cover. So we have quite a number of insurance policies running in the sector.
Loïc: It is a discussion coming up very often in investment committees and board meetings, whether we should continue to have this political risk insurance in our funds because, well, it’s also quite expensive. So we can also assess risk/return of these insurances. Let’s say that, yes, we would rather recommend to have this political risk insurance, and I also see quite some interest from the other fund managers now – not only Belgians, but also from other countries like Germany, Switzerland – in these types of policies. Certainly after the Nicaragua case, this is an issue that has received more visibility in the industry. You know, Mac, we were the only fund managers to have this political risk insurance and it was a bit strange because there are quite some amounts being invested in those countries and now there is clearly more interest from other fund managers to have this kind of risk insurance. So I think there is quite some potential.
Mac: Do you see a role for PRI in structured finance for MFI projects?
Loïc: For the moment there are not many structured finance transactions in microfinance anymore after the CDOs that were issued in the years 2005-2006, something like that. Some of them experienced a lot of trouble. But they typically have these structures with first loss layers which are at charge of the development finance institutions, of public finance institutions. But what I perceive now in the post-Nicaragua era, PRI is certainly an issue. And in future structured finance transactions, this will certainly come on the table.
Lieven: You might, for example, attract a different type of investor in having a fund with PRI cover or without PRI cover. Because nowadays, I think the funds now have a lot of development finance institutions in it or multilaterals or whatever or banks very much in the industry, but if you start to attract investors that are looking at this from a larger distance, PRI might be an interesting tool to attract the investor that seeks certain protection.
Loïc: Yeah. That’s certainly true. Of course it has to be at competitive rates…Lieven…? [laughs]
Lieven: [laughs] We always do our best here.
Mac: Loïc, what happens if you cannot get coverage in a particular country under your global policy? Do you seek cover elsewhere, or go uninsured?
Loïc: First we try to understand why the country is off-cover for ONDD. We’re always a bit frustrated if the country is off-cover because it is our mission to invest in exactly these types of difficult countries like Congo or Haiti, et cetera. We have one fallback possibility, that we can go to our government which has a special guarantee fund for microinsurance, but it’s quite limited in capacity. So we have insured a number of countries with them – among others, the Haiti investment, I think we also did the Congo with them. But as a general rule, if a country is off-cover for ONDD and we don’t find an alternative, normally we don’t invest. And it is very rare that we’d invest in a country that is off-cover, unless we are very aware of the risk that we are taking.
Lieven: To add to that, I think most of the cases where we have to say no is because of transfer problems, because our risk analysts are stating that we cannot take the transfer risk for medium-term deals. I think the latest question was on Tajikistan, and there we feel that in some cases insureds are saying “OK, let’s do it maybe one time or for a limited amount or for a shorter tenor,” but we have the impression that if we say “OK, we don’t feel comfortable with it, we cannot do it” that it becomes difficult for the insured to proceed, to convince board members.
Loïc: Yeah, and we have always the impression that the parameters that are being used by ONDD to declare a country off-cover are…well, let’s say not always very applicable to microfinance. If Tajikistan is off-cover this is based on, of course, very detailed study and I’m sure it’s very serious, but we see in the field and on the ground that there is no transfer problem, that money continues to flow out of the country without any problem, that payments are being made on our accounts without any problem. So there are other mechanisms…and these are purely…these are regular…
Mac: You’re not buying diamonds!
Loïc: [Laughing] No, no, exactly! So sometimes I must admit that we are a bit frustrated when a country gets off-cover and we say “No, no – there is no problem in that country at all.”
Mac: OK. What does ONDD consider particularly important when evaluating an application for MFI coverage? What do you want to see the investor doing in terms of risk management?
Lieven: With portfolio cover, it’s a mentality switch in the sense that the investor can add new transactions which you are not aware of. So you’re not doing the follow up on a deal-per-deal basis anymore. And there you have to trust – and this is again, really a matter of trust – that the company in question is a professional company, that what they are doing makes sense in terms of assessing the MFI, not doing crazy things, not offering something that is beyond international practice, and doing a professional job. In other words, I think we would be very reluctant to offer portfolio cover for an insured that we do not know and that’s our classic approach: we start with an insured on a case-by-case basis to see how they are doing things and in the next stage we move to the portfolio system.
Mac: Loïc, what’s Incofin’s approach to managing political risks, if you can comment on that, other than just buying insurance?
Loïc: Of course we will use typical ways of diversifying our risks among countries. Every investment proposal includes an assessment of the political situation of the country based on country studies. And then also on the ground we try to keep good contacts to the financial regulators in different countries – we know a lot of financial regulators try to understand what we are doing, that we’re not trying to take advantage of the situation but that we’re really investing in favor of the country itself. We’re also pushing a number of financial regulators to take measures, for instance, like making the use of credit bureaus compulsory. We are now involved in such an action in Cambodia, where we think this is missing in the Cambodian landscape and this is a country that might face over-indebtedness in the future and therefore we are a bit preoccupied and we are pushing the financial regulator to make this credit bureau compulsory. And there are some countries in which we just do not invest because we consider the political risk as just being too high, even if ONDD would cover them. ■