This post is part of an occasional series highlighting a project finance article or news item from the past. It is often interesting and thought provoking to look back on these items with the perspective of months, years or decades of further experience.
With this installment, we turn to a transcript that first appeared in the November 2006 issue of the Project Finance NewsWire.
Financing Projects with Unproven Technologies
Many new projects are coming to the market for financing after a lull of several years. The next wave of new construction is not like the boom years of the 1980s and 1990s when most large transactions in the project finance market involved power plants that burned either natural gas or coal and used proven technologies. Many more technologies are competing for attention in the current market. Projects that involve new ways of making transportation fuels or generating electricity or that rely on equipment that does not have a long operating history can be severely challenging to finance.
Four veterans of the project finance market discussed the challenges facing such projects in October. The panelists are Herb Magid, managing partner of Energy Investors Funds, a group of six private equity funds that has been a source of capital for many smaller project developers, John McKenna, managing director of Hamilton Clark & Co., an investment bank that helps smaller companies raise capital and list on the AIM market in London, Jerome Peters, senior vice president and group head of project finance for TD Banknorth, N.A., a prominent lender in the renewable energy and biofuels markets, and Paul Ho, director of global energy at Credit Suisse, which has been acting as the financial adviser on many innovative financings. The moderator is Keith Martin with Chadbourne in Washington.
MR.MARTIN:We are talking about whether projects with new technologies can be financed and, if so, how. Herb Magid, what is “technology risk”?
MR.MAGID:Technology risk is present in the power sector in any project that incorporates an unproven system, whether it is a turbine, a new fuel handling system, or an uncommon type of fuel. For example, there is technology risk in a project that uses a new gas turbine. The turbine may be just one component of what is otherwise a traditional power plant, but the turbine is so important that the entire plant would have technology risk.
MR.MARTIN: So it is both whether the equipment works and whether it works with the particular fuel. Paul Ho, can technology risk be addressed by focusing on the risk only during a limited time period? For example, is it enough to persuade the construction contractor or equipment vendor to guarantee that the system will pass performance tests at the end of construction?
MR. HO: When people think of technology risk, they think of it primarily as risk during the construction and start-up period. To me, technology risk is first the issue of at what level of capacity is the project capable of working after construction. Once you get past construction and start up, there is also the issue of conversion or efficiency ratio.
Lenders may be able to assume part of the conversion risk if an experienced independent engineer can certify that the plant is able to operate at least at a level that will permit repayment of the debt plus some margin for error.
A lot of energy projects, especially in the alternative energy space, involve new technologies. The general belief is that once these kinds of new projects have run successfully for a couple of years, they should continue to function properly for a much longer period of time. As a result, people tend to focus more on technology risk during the construction and start-up periods than during the operating period.
MR.MARTIN: John McKenna, is there anything you would add to the definition of technology risk?
MR. McKENNA:Technology risk related to products, for example, might be the risk of whether the piston ring will work and whether the mechanical engineering is correct. It is very difficult to get an engineering or construction firm to provide a “wrap” guarantee of this process technology. Technology risk is a question of whether the developer can prove that this particular product will work over time.
MR.MARTIN: Jerry Peters, isn’t there a degree of technology risk in every project?
MR. PETERS: Yes, to one extent or another. Many technologies have been demonstrated over a long enough period of time to make lenders comfortable about the risk, but the reason lenders analyze the debt-service coverage ratios in projects is because there is an element of project risk that never disappears completely.
Lender Risk Tolerance
MR.MARTIN:What degree of technology risk presents a challenge for raising financing? How do you know whether a particular project presents too much risk?
MR. PETERS: How many successful projects use the same technology? Unless a fair number of projects using the same technology have been in operation for a period of time, then you have an element of technology risk that you really cannot assess. It is when the technology risk is unquantifiable that I think you have reached a level of risk that a lender usually cannot absorb.
For example, in a conventional power plant that uses a combustion turbine that has many hundreds of thousands or millions of hours of operation, you pretty much know that it will operate within 95% to 100% of its projected output and efficiency simply because other, similar turbines have operated that way. If you are looking at a type of cellulosic ethanol plant that has never been built, you really don’t know
at what through-put efficiency level it will operate. At that point, you say the project has an undeterminable level of technology risk. Therefore, I cannot take the risk.
MR.MARTIN:What should someone who is bringing the first project of its kind to you bring with him to prove the technology risk is manageable, or is it just impossible?
MR. PETERS:We are dealing with that question with some of the biofuels technologies today. It comes down to whether you have already demonstrated that the technology works on a smaller scale. Is there a successful pilot plant? Can you then get a construction contractor to take the risk that a technology that works in a pilot plant will work when it is replicated on a commercial scale? The construction contract must have adequate through-put efficiency and time-delay liquidated damages to repay the debt if the project does not work as guaranteed under the construction contract.
MR.MARTIN: So the maxim in project finance that lenders won’t accept technology risk is true?
MR. PETERS: Pretty much.
MR.MARTIN: Paul Ho, must the construction contractor take the risk?
MR. HO:The construction contractor or the equity provider should wrap the risk. If you have a technology that is being championed by someone with a substantial balance sheet, that champion should, in theory, be willing to stand behind the technology.
MR.MARTIN: Herb Magid, if someone comes to you as a private equity fund with the first project of its kind, would you consider putting money into the project?
MR.MAGID:We have looked at a lot of first-of-its-kind projects. Those are the kinds of deals that usually have an individual developer who is knocking on all of the doors, and they are often the most interesting projects to think about. We have invested in the past in several early-stage development projects. One was a tire pyrolysis plant in London that made so much sense on a variety of levels, and the components of the system were all proven. The risk was in combining it all together. At the end of the day, as Jerry Peters said, the construction contractor was unwilling to wrap the final performance, and the project was never able to get financed.
We continue to look at such projects, but a project has to be more along the line of one of the first 10 GE new turbines, with proper guarantees and reserves, as opposed to a brand new system. I view the brand new systems as more of a smaller-scale venture capital investment than a large project finance transaction where you are relying on 20 years of cash flow.
MR.MARTIN: I was wondering whether an equity investor looks at this differently than a lender. The answer is the equity investor wants to know the lender will be there eventually to finance construction.
MR.MAGID:That’s right. Unless you are going to fund the project entirely with equity, you will need eventually to bring in lenders and have the proper wraps. If you don’t think you can get there, then it will be hard to persuade a private equity fund to provide development capital.
MR.MARTIN: John McKenna, you have spent a great deal of time thinking about how smaller ventures get off the ground using new technologies. We have now heard from both the lender and the private equity investor that the lender will not take the risk on unproven technologies and the private equity investor won’t provide funding unless he is confident a lender will eventually finance construction.What would you advise someone whose project uses a new technology? How should he start moving down the path toward financing?
MR. McKENNA:That is precisely the dilemma. If Jerry Peters can’t afford to take the risk, then the project has a problem. All of us who have been in the lending business know that the profit margins for banks are much too small to accept that kind of risk.
The private equity venture capital world for energy technologies today is rather small. The average deal size is $5 million to $7 million, and that is nowhere close to the kind of equity capital that is required
for many projects.
We advise clients to look at other alternatives. Try the larger private equity firms. I was in New York for the Private Equity Analysts conference a couple of weeks ago.That’s the crowd: non-traditional energy
tech venture capital firms. Also, the London Stock Exchange has been the major source of equity for some of these larger project-type companies.
MR.MARTIN: The London Stock Exchange meaning the AIM?
MR. McKENNA:The AIM portion of the London Stock Exchange.They have served as a source of pure project equity financing. Yes, the London AIM is another alternative.
MR.MARTIN: I heard from Herb Magid, who runs six private equity funds, that even for him it is tough if he can’t foresee the lenders putting money up to construct.
MR. McKENNA:The developer must prove that he has all of the other components in place, like a proper wrap from a construction contractor or backstop from a sponsor with deep pockets. It is also a question of risk-reward ratio.What return is he offering the private equity shop in relation to the risk he is asking it to take?
MR.MARTIN: Paul Ho, I know you have spent time thinking about how to finance novel projects, ones that may not be the first of their kind in the world, but perhaps the first in the United States.What advice do you have for someone who is trying to build such a project?
MR. HO: I echo some of the other comments. I think the technology risk for an unproven technology that has not been proven on the pilot scale is too great for project financing. Such a project will have to be financed primarily with equity. Once such a project has run successfully for a few years, then maybe some of the equity can be replaced with project-level debt.
Another possibility is perhaps you can separate the part of the project that has the greatest technology risk from the rest of the project, and use traditional project financing for the part that does not have the heavy technology risk and all equity for the part that does. That is the strategy we are pursuing for Rentech, a US coal-to-liquids developer for its first commercial-scale project. Rentech purchased a fertilizer plant in Illinois that it intends to convert into a coal-to-liquids facility. The technology risk is for the Fischer-Tropsch portion of the project, or the part that takes gas made from a gasification train and converts it into a liquid fuel.What Rentech plans to do is to treat the FT portion of the project as a separate project and finance it with equity. The gasification train and fertilizer plant can be financed with traditional project financing. Overall, the project will end up with a blend of equity and traditional project financing.
MR.MARTIN: Herb Magid, I thought I heard you say — or maybe you merely implied — that you would not put in all equity for a completely new technology?
MR.MAGID: I suspect Paul is describing a case where the equity will get the benefit of two potential cash flow streams — one proven from fertilizer sales and the other more uncertain from the FT liquids sales, but like any equity, it will stand in line behind the debt. We are looking at a lot of coal-to-liquids and IGCC plants, and it is really a challenge to figure out a way that you can make an equity return with reasonable risk given the huge scale of these projects. It is hard to see how such plants will get financed without someone like the federal government or a deep-pocket construction contractor stepping up and guaranteeing performance.
MR.MARTIN: So once again, the project is not going to be able to secure full funding from a private equity fund. You are still looking to have a lender alongside you?
MR.MAGID:You really have to, unless you go down the tax side where some new technologies can be financed with all tax equity, and then there is less of a concern whether the technology actually works because the tax benefits are a function of construction cost rather than output.
MR.MARTIN: Jerry Peters, you said that one way to get past technology risk is to get the construction contractor or the equipment vendor to agree to take it and pay suitable liquidated damages. How much in liquidated damages and over what time period of coverage would you require?
MR. PETERS: You have to break that down into several components. First, you can’t limit the liquidated damages to anything less than the complete construction cost until you get to mechanical completion, because you have to make sure you have a plant that has been fully constructed. The plant has to be guaranteed to reach the stage where it is ready to start up. If the contractor guarantees you anything less than that, the risk exposure is too great.
Next, we have to move the plant from mechanical completion to substantial completion by demonstrating that the project can work. You must set a target for what the plant can do. For example, in a biofuels plant, the target is 50% of capacity. The usual limit in the industry for liquidated damages during that period is 10% to 20% of construction cost.
Next, you have final liquidated damages for the period when the plant is moving from substantial completion to final completion. During this period, the plant is tested fully to prove that it is capable of operating at the minimum capacity promised in the contract. The level of liquidated damages does not usually drop below 10% of construction cost during this stage.
If you can get those levels of damages, then we would generally be comfortable.
However, if you are talking about a project that is a scale up from a pilot plant, then we may need to maintain the guarantee at the mechanical completion levels through final completion because there are a lot of variables involved in start up that may not be able to be proven at the mechanical completion date.
MR.MARTIN:Most of what you said has to do with completion risk. Are you not as concerned about operating risk? Are you not as concerned about the risk of technological obsolescence during the 10 to 12 years that the debt will remain outstanding?
MR. PETERS: It depends on the technologies that you are employing. A lot of us found out when we were financing some of the newer models of combustion turbines that the improvements in efficiency that were claimed by the manufacturer did not pan out over time. If you are looking at a new technology where a 3% or 4% difference in efficiency could mean the difference between a project that provides adequate debt service than one that doesn’t, additional performance guarantees may be required.
Focusing again on biofuels, once you prove a conversion ratio — once you prove energy consumption — there is not much about the process that will go wrong because it is a very simple process. It is a lot of tanks and pumps and gears and things that generally don’t go bad.
Everything is technology specific. The length of guarantee required varies potentially with each technology. It is very, very difficult to get a construction contractor to provide more than a one-year performance guarantee on the process engineering. Obviously on the wind turbine side, we have several manufacturers will give guarantees lasting up to five years. The length varies with the technology involved.
MR.MARTIN: John McKenna, suppose a developer has a process for gasifying biomass or poultry litter, and he is planning to use a process that has not been used yet in the United States. The normal approach for this type of fuel is to burn it directly. This developer wants to gasify it first.What would you advise him to do before approaching Jerry Peters or Herb Magid?
MR. McKENNA: I don’t know if you asked me that question because you know that I spent three years trying to develop a company that actually tried to use chicken litter in a Stirlingcycle engine. I spent three years of my life with that technology. The answer is that the technology really has to work, and it has to work for a long period of time. It has to have a warranty reserve, which is almost a working capital reserve for the company. It has to have worked at scale before Jerry is even going to agree to a meeting. He is going to ask me these kinds of questions over the phone before he will agree to meet. A company in this position has little choice but to try to raise equity capital, thoroughly test the product and then look for debt financing.
MR.MARTIN: You said venture capitalists will put in money for early stage development. How large an investment will venture capitalists make typically? What is the maximum?
MR. McKENNA:The trends we are seeing now are in the $25 to $50 million range in terms of total amount, and that probably assumes four or five venture firms will invest alongside one another.New process technologies will probably need $50 million to get through the beta stage. It is during the beta stage that the testing regime occurs.
MR.MARTIN:“Beta stage” means what?
MR. McKENNA:The beta stage is where the equipment has worked for X years or X period of mean time between failure (MTBF). The original question was how technology that will end up going into a poultry-litter-to-electricity plant would make its way through the testing process so the plant can get financed. I believe that venture capital equity has to finance it during the equipment testing phase. It is only after the project has moved past the beta stage that Jerry will organize the meeting with the developer who is using this new technology.
MR.MARTIN:Why are the venture capitalists more willing to put in money than Herb is as a private equity fund? What sorts of questions would they want answered before they will invest?
MR. McKENNA: One question as technology moves into the beta stage is who financed it during the alpha stage and what performance has been promised? What is the mean time between failure for that particular product? What is the overall return for this business? Where will this business end up going? How quickly can the venture capital investor see an exit or assure that the B and C rounds of financing get the technology to commercialization? I would say today that maybe a 25% to 35% internal rate of return will be required for this kind of equity.Herb,would you agree?
MR.MAGID: Yes, I would. I think the big difference is that the venture money is looking for a quicker and larger return, a way of taking this technology and building a company and getting value for selling more units, where the traditional private equity investor in a power or other infrastructure project is looking for a return over a very long period of time. The technology risk continues. It is not just construction and start-up risk. It really is a long-term operating risk, and you do have to have the reserves and the comfort that the sponsor will have skin in the game — a back-end interest or a carried interest — that will make him keen to see that the financial projections are met. There is a big difference in what a venture capital investor and a more traditional private equity investor will require. This then affects the type of diligence that each will do before making an investment.
MR.MARTIN: Jerry Peters, if the developer who is planning to gasify chicken litter can show that there is one other plant that was built recently in Europe, and it is working properly, is that good enough to get over your hurdle on technology risk?
MR. PETERS: If the technology has not been demonstrated in any other place other than that one plant, I would probably find it difficult to finance the project. A lot of the problems specific to that technology have to do with corrosion, and that is a longer-term problem that you can probably address with performance guarantees from a construction contractor or equipment vendor.
With that particular technology, if there is only one plant in operation — whether it is a demonstration plant or even full-scale facility — I would have trouble making a loan if the plant has not been operating for at least the period of time for which the developer wants to borrow money.
MR.MARTIN: At what scale must a pilot plant have worked in relation to a commercial scale plant before you will accept that the technology works?
MR. PETERS: The answer is technology specific. If it is a modular technology, as in gasification, you just add modular gasifiers; then the scaling up of the plant is easy.
An example is where the pilot plant has one gasifier and the new plant will have 10 gasifiers. If it is another technology that is not modular and no one has built a certain component that is needed for the commercial-scale plant, that would present a level of technology risk that I would probably be unwilling to take.
MR.MARTIN: John McKenna, you mentioned there are just a handful of venture capitalists in clean tech.What is the best way to identify those people? How do you find them?
MR. McKENNA:There is a Cleantech Ventures organization of all venture capitalists that invest in this area. This includes energy tech plus other clean technologies. The National Renewable Energy Laboratory organizes a conference every year that is very well attended by all the major players in the industry.NREL has a website that includes a comprehensive list of investors.
Other Ways to Allocate Risk?
MR.MARTIN: Paul Ho, suppose we got past the venture capital round and managed to persuade Herb Magid to invest some equity and Jerry Peters is willing to consider lending, but he doesn’t want any of the technology risk. You are a project finance expert, and project finance is an exercise in deconstructing and parceling out risk. There are a number of potential risk takers. Is insurance an option? What about making the offtakers, sponsors, construction contractor or technology licensor each take a share of the risk? Where would you try to put the risk? What has been your experience?
MR. HO:That’s a good question. I agree with Jerry Peters. The number one guy to wrap this risk is usually the construction contractor. If he is not comfortable, we would look next to the equity provider. Sometimes in very rare circumstances you might have an offtaker who believes enough in the technology that he might be willing to enter into a tolling type agreement and take the conversion risk, but usually subject to some minimum thresholds.
MR.MARTIN: Construction companies are feeling beleaguered. Everyone wants to place the risk on them. Are there construction companies that are willing to take this risk in practice?
MR. HO:You would think that construction companies are in a better position than almost anyone else in the deal to evaluate the risks, especially contractors who deal regularly with the big-name technology providers and who partner in large-scale projects all the time. For example, with coal gasification projects,we will look for someone like a Fluor or Bechtel to work with a General Electric or ConocoPhillips. The two parties together will provide a comprehensive wrap or have the contractor alone provide a wrap on the technology, and in turn work out a back-to-back guarantee with the technology provider or equipment vendor. From a lender’s perspective, that is a much cleaner approach than having different people guarantee different parts of the construction.
MR.MARTIN:What about insurance — have you seen it used effectively?
MR. HO: I personally have not seen insurance used effectively in wrapping new technologies.
MR.MARTIN: Jerry Peters, have you seen insurance used?
MR. PETERS: I have seen it tried several times, but when you drill down into the insurance coverage itself, you find out there are a lot of gaps. There is a big difference between having a performance bond on a construction contract, which is pretty easily collectible, versus trying to get an insurance company to pay under an insurance contract. There are usually many more outs for the insurance company.
I have been in this business for 25 years, and I have never seen a federal loan guarantee get done either.
MR.MARTIN: I was going to move to that.Why have loan guarantees at the federal level failed?
MR. PETERS: I think the current two-billion-dollar-level guarantee program run by the US Department of Energy is a good example. It will fail because the guarantees have been structured to put the lender in the first-loss position.The government guarantees up to 80% of the debt amount which, in turn, cannot be more than 80% of the project cost. The US government takes a first lien on the project.The lender is left with a second lien with respect to the 20% of his loan that is not guaranteed. This leaves the lender in the firstloss position. We would not be comfortable taking technology risk in such a financing structure. In this sense, the program is a failure. It was supposed to encourage lenders to finance projects that use new technologies. As a lender absorbing the first loss, that is the very last thing we would want to do.
MR.MARTIN: Herb Magid, have you had experience with government guarantees?
MR.MAGID: The only one that worked for us was about 20 years ago, and it was on a geothermal project where we were already fairly comfortable with the technology risk so the guarantee merely served as additional comfort. I think Jerry is right. If there is a meaningful risk that guarantee will be needed, then you really have to look to the equity or reserves to take the first hit, but not the lenders.
MR. PETERS: Let me be clear. I have not said the guarantees are worthless. I just have not seen deals done with them.The guarantees could be worth something if you can find a way to structure around that 20%, that first loss on which you are sitting with only a second lien. If we can structure around that, I would be more than happy to have the federal government guarantee 80% of my loan.
MR. HO: Are you saying that you would be happy with a pari passu first lien on the 20% unguaranteed piece?
MR. PETERS:That would be one way. Another way — again, currently prohibited — would be to do strips. Paul, you might be more than happy to do that 20% strip and then get real low-rate lenders to do the government guaranteed strip.
Will the government allow us to do strips? I don’t see the harm to the government of allowing them.
MR.MARTIN:“Strips” meaning one lender will lend at a higher rate for the first-loss portion.
MR. PETERS: I am more than happy to sit with the government-guaranteed piece and get 80 basis points over.
MR.MARTIN: You narrow the technology risk, but somebody is still going to have to step up for it.
MR. PETERS: If you look at the weighted average cost of funds where you have one lender that is taking the second lien piece and the 20% percent uncovered risk at perhaps 1,000 basis points over, and then you combine it with an 80% guaranteed strip that is done at less than 100 basis points over, you still have a very good capital structure for the leverage in the deal.
It is very difficult to get senior lenders that are accustomed to taking project risk to take that 20% first-loss position. There are lots of funds that have no problem with taking that position.
MR.MARTIN: John McKenna, have you seen other sources of funds, perhaps state clean tech funds that people might try to tap as well as venture capital?
MR. McKENNA:Well, I think that the big source of capital in many of these projects — and this is especially true of projects that produce liquid fuels — will be the oil companies. At least in the biofuels area, until there is a buy in by the large oil companies with the capital to invest in these kinds of projects, the market will never reach its full potential. Is there grant money? Yes, there are some small grant programs, and there are an incubator programs organized by the US Department of Energy. Details can be found on the DOE website. These are relatively small sources of capital, but nowhere near the amount of capital required for the projects we are talking about.
The only other source that we have not discussed is strategic investor money, which is primarily from energy companies with a strategic interest in seeing the technology develop.
MR.MARTIN:To sum up, it seems like for projects that use a technology that has not been proven, and “proven”means used more than once successfully, you are basically talking venture capital. Once you get past that stage, you are still going to have to show someone like Herb Magid or Jerry Peters that the technology risk is covered by somebody else in the deal besides the lender, and that somebody else is probably the construction contractor or the equipment vendor.
Although, in theory, it is possible for others to step forward, like the government or insurance companies, government guarantees and insurance have not yet evolved to a stage that adequately covers the risk.
MR.MAGID: Keith, one potential source of support that is probably worth mentioning is the various state agencies. For example, the Massachusetts Renewable Energy Trust has a pot of money that has grown each year through a surcharge on electric bills. It was $150 million the last time I looked, and the trust is trying to do creative things to support renewable technology and efficiency. This is not the same thing as covering technology risk, but it helps to have in place price supports to renewable energy credits. Again, this support on a smaller scale, but some states recognize that seed money and support are required if some of these new technologies are ever going to get off the ground.
MR. PETERS: I want to go back to the construction contract side of things because,Keith, you mentioned they may be feeling a little beleaguered. In certain technologies, there is a huge potential for growth — one of those being the cellulosic ethanol field — so that there are enormous gains to be had by any construction contractor who is a first mover.
If you look at some of the early builders of ethanol plants, like Fagen Inc., the fact that they held licenses to a technology that was eventually proven gave them first-mover status. They could then deploy the technology by duplicating it in plant after plant.They have made an enormous amount of money. I suspect other construction contractors are looking at how they can get first-mover status in other areas. The point is contractors have an incentive to provide guarantees to insure that a market in which they will enjoy first-mover status can get off the ground.