CMBS Debate: Getting real estate financing right

Roundtable participants

SB, Euromoney  How has the development of a CMBS market changed real estate finance?

CC, Deutsche Bank  CMBS, in our view, is not so much a product as a tool which can be used as an overlay of the actual product we banks have to offer, which is a real estate loan. CMBS is just one of the exits we can use and I think sometimes people forget that distinction. Two other terms that are commonly used but also similarly misunderstood are "conduit" and "securitization". Conduit, as we would understand it, refers more typically to what is understood in the US as the small loans business. This involves typically securitizations of pools in excess of 50 loans. By contrast, in Europe we have tended to think more in terms of the pooling of fewer, larger loans, which we normally refer to simply as securitization.

SB, Euromoney  So what has driven the recent upsurge in interest in CMBS in Europe and how does the market's evolution differ from what happened in the US?

CC, Deutsche Bank  The advent of the securitization market in the US was relatively sudden, driven by the lack of funding available for most forms of real estate that followed the S&L collapse. In Europe the syndicated loan market has historically not suffered a similar crisis, which has meant CMBS volumes have developed more slowly as an alternative source of liquidity for what we do day to day – real estate-based financing. The growth we have witnessed in Europe was originally driven partly by increased return expectations which pushed up required loan margins. This was also driven by a changing regulatory environment (Basle II), the end of state subsidies and the end of state guarantees, forcing banks to price real estate loans at wider levels. What has driven the recent surge in volumes has been the dramatic drop-off in CMBS spreads over the last two years. Triple-A spreads have gone from 45 to the low 20s; Triple-B spreads have fallen from the 400s to the sub-200s. That has given the CMBS market a competitive price advantage and that, combined with the change in the regulatory environment and the increased return requirements of portfolio lenders, is what has driven growth.

Maintaining relationships

SB, Euromoney  Are borrowers used to the loan markets anxious that if they use securitization they will lose the close relationship with the lender that they rely on to work for them if times get tough? Do they find disintermediation and the thought of dealing with dozens of bondholders a threat?

CC, Deutsche Bank  The answer to that is that we are real estate lenders, and if we cannot address questions of this kind from borrowers then we don't have a sustainable business. One of the distinctions between the market in Europe here and the market in the US is that this is one of the critical components. If we cannot address that up front, it doesn't matter what margin, what fees, what leverage we provide, there's no business. All of our business remains managed and serviced by people within the bank and we do not outsource any of our bond or loan administration. PC, British Land  I agree with that, but as a repeat issuer you do also look through that relationship to who's holding the bonds, because you also have a relationship with bondholders.

CC, Deutsche Bank  That's right, and clearly lenders like us do have a very strong relationship with our bond investors in Europe and in the US. And if there is a request from one of our borrowers which we support, we will typically obtain approval for it on behalf of our borrowers.

JB, Lehman  It's a similar role to that of lead agent bank on a syndicated loan. You do everything you can to get the syndicate banks to cooperate with the borrower. Canary Wharf is a good example. It's a company that's undergone a lot of change and there were a lot of times we had to talk to the rating agencies and to investors about getting things approved. In my experience, the securitized lenders are there to serve the client and get the deal done and then live with that deal as it goes on.

RG, Morgan Stanley  That's absolutely right. Cyril and Jim are excellent lenders for a lot of reasons, but one of them is their cooperative attitude. It means we don't have to be bond experts; it means we don't have to call the trustee alone when we need modifications because of changes to our business plan; and it means we can get their help with the flexibility we need, even if it wasn't contemplated up front.

JB, Lehman  Most of us around the table think of ourselves as secured lenders. So first we deliver what the client needs and then we ask whether that loan works better in the syndication market or the securitization market? At the end of the day, the borrower wants a loan to accomplish their needs, and then it's up to the lending side to figure out how best to finance that loan. The big shift has been that securitization used to be a tool for very sophisticated borrowers, with very large financing needs and long time horizons who could not find that financing in the bank market. Now that the structures satisfy the needs of a wide range of borrowers, and not just a few big guys and the ratings agencies, the market has really taken off.

SB, Euromoney  Zubin, as a borrower, what has attracted you to the market?

ZI, Goldman Sachs  When I came here six years ago we basically did all of our financing through the bank market. In the last 18 months to two years that situation has reversed and now almost all of our financings are securitized and I would say that for large, complex financings securitization is almost the only option. Also, where you have long leases with credit tenants it certainly makes sense to securitize. Where we find securitization to be difficult – other than rating agency constraints – is on the hedging side. For securitization to get the best execution you have to do a collar or swap – something other than a cap – and we find that the most challenging part of it. Prepayment penalties are also an issue: we're IRR players and so we want to exit assets as soon as we can. When there are prepayment penalties – which is necessary for the banks to make money on these securitizations – that can be difficult.

SB, Euromoney  It sounds as though the biggest issue, though, is size – borrowers are being pushed into the securitization markets because deals are getting bigger?

ZI, Goldman Sachs  Yes. For quite some time the biggest deal we did was ENI at $1.2 billion. These days that is a medium-sized transaction.

PC, British Land  I agree with Zubin. If I look back at why we really started in the securitization market, it was a question of capacity. When we launched our Broadgate securitization in 1999, we could not have raised nearly £1.5 billion from a single source. Nor could we have raised that sum from the various other sources available to us in a reasonable timescale. We now have over £12 billion-worth of assets, and £6 billion of borrowings, and I don't think that would have been possible without having access to the securitization market. So I think the primary driver for us was that capacity issue: where else do you go to get the money?

RG, Morgan Stanley  And even if you can get the capital today, large, regular borrowers will eventually meet concentration limits with lenders. With securitization you can build up a relationship with a lender and avoid them running out of capital to allocate to you because they constantly replenish it through the capital markets.

PC, British Land  Exactly.

AR, NM Rothschild  It is interesting though that if you look at the securitization market over the past two to three years, there are two extremes. There are the very big ticket deals – the Broadgates and the Canary Wharfs – and then the multi-loan conduits – initially, the Anglo-Irish and building society deals – where the borrowers did not know that they were participants in the securitization market. What we've seen is a slow closing in of the two, where the potential for smaller borrowers or those with smaller portfolios of loans to access the market has increased considerably. I also think that borrowers – and the investment banks – are a lot smarter about how the whole capital structure is priced – investment grade, mezzanine, junior and that has meant that commercial banks have had to move to a more transparent market pricing.

JB, Lehman  Very few borrowers look at the capital markets exit and try to negotiate pricing on that basis. Most have a financing need and if you can fulfil it at a price that makes sense to them then that's it.

CC, Deutsche Bank  It's real estate lending. The borrower's concerns are price, leverage, security minimization and constraints. There is a certain category of borrower who tries to reverse-engineer the exit and the pre-payment penalties and so on but that kind of borrower exists on a very specific type of deal. The majority have other concerns, such as their business plan, their acquisition structure and getting the best financing available.

Relationship versus disclosure

SB, Euromoney  We've talked a little about the fact that CMBS do not change the basic lender-borrower relationship. But isn't one benefit of loans versus securitization that disclosure is less – bondholders are demanding levels of disclosure that borrowers are uncomfortable with?

JB, Lehman  We would like a bond that we issue to be tradeable until maturity, and that means that the deal must give you the information that you need to trade it. That's one of the things we worry about on the agency side.

CC, Deutsche Bank  That's an important point. One of the few areas where there exists a difference between securitized and syndicated lending is disclosure. It is possible to restrain the amount of information that's distributed to the lenders on a syndicated loan, especially if it's a club deal or the transaction is relatively small – say £100 million to £150 million. The CMBS market requires full and detailed disclosure and we have to provide that without breaching the confidential details of our clients' business. We work on the basis of detailed disclosure with our bond-holders, subject to confidentiality agreements and other restrictions.

AR, NM Rothschild  Yes – though the secondary market poses much bigger challenges in terms of the availability of information.

PC, British Land  I think there is definitely a difference between the disclosure we believe to be best practice and the detail being demanded by the securitization market. We've had pressure from investors saying: 'We want a full quarterly cashflow on a tenancy-by-tenancy, lease-by-lease basis'. Now that information is of course commercially sensitive.

JB, Lehman  The market needs a compromise. In some of the earlier conduit deals it's very difficult to trade the bond because the offering circular doesn't really tell you what properties were in at the start or are there now. Right now I think disclosure is pretty good.

Evaluating alternatives

SB, Euromoney  Steve, you've not embraced securitization.

SO, Brixton  No. We've gone down the unsecured route. We converted all our bank debt in 1992 from secured to unsecured; then we tapped the sterling bond market in 1999, and now we have £485 million of investment-grade unsecured bonds in issue and we have total flexibility on balance sheet for what we can do. We can hedge however much we want; there are no conditions in our facilities as to how much we can hedge; since 1997 we've turned over about £3 billion-worth of property, and last year alone we turned over £1.5 billion. CMBS has been interesting because in the last year pricing has come in so much. But if you take into account all of the costs – hedging and constraints – what is the total cost to a borrower as opposed to the headline margin? Unsecured banking market margins have dropped for us this year and we can borrow at about 40 to 45 basis points with complete flexibility. And given that base rates move in 25 basis point shifts, when you've got that flexibility on hedging as well, over the medium term how much are you likely to save in terms of funding costs, compared with the flexibility? We also have two joint ventures and there we have secured, non-recourse club facilities with the same banks that lend to us on balance sheet. Many of those banks are also in the CMBS market, so it will be interesting to see within our joint ventures whether a CMBS structure ends up replacing a traditional five-year secured club facility. Bond investors have been asking us about this since British Land securitized Broadgate in 1999 and spooked the sterling bond market, because everyone thought the whole property sector was going to securitize. Given the nature of our portfolio, which is industrial property, where lot sizes are much smaller – it's very unusual to have a lot size over £50 million – and given our churn, I don't think securitization is going to be appropriate for our business. We've got complete flexibility, and pricing in the sterling bond market is being driven down by investors looking for yield and the fact that there are only three quoted companies in the property sector issuing them. So that puts pricing tension in our favour.

ACh, Euromoney  In the unsecured bond market there's been greater pressure for stronger covenants. Have you found that an issue?

SO, Brixton  No. In our last bond issue we put a change of control clause in – a put at par – and we were quite happy to do that as it benefited us in terms of pricing but left us with complete flexibility. For example, the outer borrowing limit on our unsecured bonds is a 175% debt equity ratio, and even in the bear market of 1992 our debt/equity ratio was 114%. It's very, very flexible. PC, British Land  Do you have an interest cover covenant?

SO, Brixton  On the unsecured bond, no; on the bilaterals yes.

RG, Morgan Stanley  There are a basket of considerations that come along with covenants. We like to think that our 14-year history of investing in properties and performing well by our lenders and note-holders will be an increasingly important differentiating factor among lenders and bondholders. The covenant package is important, but a borrower's history, despite the fact that these are non-recourse loans, will have an important weighting with the note-holders.

AR, NM Rothschild  Borrowers have to look fundamentally at whether on-balance sheet is suitable for them. They have to look at using a combination of loans and securitization. They need to look at leverage: at higher leverage, clearly securitization works up to a point and then there's a massive step change in terms of pricing. Banks are beginning to be creative in how to use their balance sheet alongside CMBS and I think it's quite rare that an entire property business would be purely CMBS-funded. It's bringing all those tools together, with CMBS as a significant part, that has driven developments.

RG, Morgan Stanley  That's definitely true at high leverage. Some banks are using their balance sheet for the mezzanine that in the US would be the double-B, single-B and non-rated pieces of paper. Those pieces are often held aside from the securitization in the form of a B-note and either held on balance sheet by the bank or syndicated to other debt investors. Usually CMBS lenders have limited appetite for holding that kind of risk indefinitely and so the distribution group within the securitization group will sell these B-notes, which might not be securitizable today, but nevertheless are sellable.

Flexibility concerns

ACh, Euromoney  So banks feel that the borrower-lender relationship is pretty much unchanged and disclosure requirements at the medium-size to large end of the market have converged. But Steve has raised a couple of important points: what about securitzation's perceived lack of flexibility relative to loans? And then what about all-in costs?

RG, Morgan Stanley  A crucial evolution has been that the securitization market now has a price for whatever flexibility a borrower may need. This was not as true three years ago, when the principal loan market was more flexible than securitization. So for us in terms of what securitization brings, the answer goes back to a point that Cyril made a few moments ago: it's more about what does the loan bring? The question is not loan or CMBS, it's what each avenue means to us and the lender. One thing securitization can often mean is that you have more constituencies to go to if you want to make a change; to make securitization appealing that issue has to be addressed and sometimes paid for. So if you know that you need to go to a trustee and to a note-holder vote to get some kind of change made, that's a steep uphill climb and you'd rather have a simpler decision-making process. But if you know in advance that, for example, you will want to separate a particular site from the rest of the collateral to sell it or finance it separately, then securitization becomes unconstraining. That said, securitization structures are becoming much more flexible in terms of the ability to substitute properties in or out, the ability to have a release premium if properties are taken out and so on. This is leading to a point where we can be indifferent about the securitization and focus on the characteristics of the loan that satisfy the underlying business plan.

AR, NM Rothschild  It's an issue quite close to the hearts of all of us here. Certainly in the past it has been a client concern. But let's look at what clients mean by CMBS being inflexible. First, it's the ability to buy and sell property. Substitution is something of a misnomer, because it rarely happens, but clients want to be able to sell a property and then use the proceeds for something else. They want to be able to change leases, so what is the consent process? And then beyond that is whether they can raise more money in the future without incurring a lot of new costs. In addition, clients want the same types of structures in CMBS as they have within loans, such as revolving facilities.

Two years ago many of these things would have been difficult or impossible. But the market has come a long way in terms of asset management flexibility – in some of the less-leveraged deals there is 100% flexibility and 20% to 30% per annum is typical in most new transactions. So clients can buy or sell property without needing to go through a rating agency approval process each time. What they still need to do, though, is ask themselves what investors will buy. Borrowers may want flexibility but investors don't want paper that will look completely different in two years' time. PC, British Land  I'm not sure just how inflexible the early structures were. We used securitization in 1999 and we wouldn't have done so if we didn't think that it gave us the flexibility that we wanted. I think the real change has been the ability to substitute assets and the movement towards rating a blind or semi-blind asset pool. This has been possible because the rating agencies have improved dramatically their understanding and knowledge of the real estate market over that time.

CC, Deutsche Bank  I would argue that there is more flexibility in a securitized structure than a syndicated structure, as long as it's built in ahead of time.

PC, British Land  Yes, but it depends what you compare it with. As Steve was saying, if you compare it with unsecured lending, where you have absolute flexibility, then it's certainly inflexible. And the interesting thing is we actually run our business in a very similar way to Steve. We keep a pool of unsecured debt as well – we've got £4.2 billion of unencumbered assets, which we borrow unsecured against. So the decision you're always making is: should we swap something which has complete flexibility for something which is by its nature more restrictive but has more advantages?

CC, Deutsche Bank  I agree with you, I think you are comparing two products which have different pricing, different limitations and different constraints as well.

PC, British Land  Except for the fact that the pricing is not very different now and the flexibility has converged.

CC, Deutsche Bank  Even taking leverage into account?

PC, British Land  But you can get leverage on unsecured up to say 65% loan to value at about 50bp over Libor. If you can do a securitization at 30 or 35 over, that's when your decision point comes in; how much are you going to pay for that flexibility?

SB, Euromoney  Peter, what were the key flexibility terms that you demanded to do that transaction and that you achieved?

PC, British Land  We were really focused on asset management. We knew that we were in a better position to manage the properties than a bank or rating agency; after all that's what we do. And so what we needed was the ability to build in accepting surrenders, granting new leases, doing rent reviews, granting rent-free periods – doing whatever we thought was best. Everything we do has to be in the best interest of the bond-holders, but it's axiomatic that if it's in the best interests of the equity-holders, who don't get paid anything until all the bond-holders get repaid, then it's in the interests of bondholders. For instance we did a deal with EBRD, where we re-geared their lease and gave them three years rent-free. We did that without having to go back to either the trustee or the rating agencies to get the consent.

SB, Euromoney  Andrew, has the legal framework got easier in the last three or four years in this market or has anything fundamental changed?

AC, Clifford Chance  Well it's certainly true that if you build flexibility into the loan at the start of the deal then you can do most of what you want. So in terms of a regular loan, I don't think there's much difference between a loan which is being originated for securitization or a loan being originated for syndication. The bigger deals, and one-off deals being done for a particular balance-sheet or tax reason, can still take months and months and they can be extremely complex. It's important not to confuse those with regular property loans which, by and large, are getting easier. I think an issue has been that the market was not set up for securitization. Legally, the biggest problem has been fitting essentially US practice and English language into the continental models. There is pressure to commoditize, but that is difficult to achieve. The fundamental thing is that the process is becoming more streamlined.

Measuring securitization's true costs

SB, Euromoney  The other big question, alongside flexibility, is cost. What is the real all-in cost of a securitization versus a principal transaction if you take everything into account?

PC, British Land  The real answer to that is time. I harboured great hopes when it took us nine months to do Broadgate that the process would become easier and more simple as the years went on. The time period has shortened a bit, but I'm not sure that the process has become any easier or any more simple. So when people ask me: 'Should we do a securitization?' one of the things I always say is, 'It's going to take up a lot of time and effort from you to get the thing done'.

SO, Brixton  Absolutely. In the unsecured market you can go out and do a syndication in the market in six weeks for £1 billion if that's the route you choose. Or like us you can do it bilaterally in around four weeks. To get a similar size for securitization, though, can be four to five months and the costs of that wait can be significant – you can miss favourable rate and spread environments by months. PC, British Land  Well you can fix your interest rate upfront fairly easily, but it's still very difficult to lock in spreads if you think they are attractive.

RG, Morgan Stanley  That's the important distinction. How many people actually factor in the risk of credit spreads widening into their deal costs? If a borrower pays 50 basis points of fees for an agented securitization all in – distribution, legal, rating agencies – compared with paying, say, 75 basis points for a principal loan which takes six weeks to closing, people are going to look at the two and compare the absolute amounts. But sometimes borrowers do not take into account the fact that they're locking in spreads and total interest rate within a few weeks on a principal commitment, versus waiting to lock in spread rates in several months on an agented deal. A borrower may have a view on how the credit of his project may evolve over the near term. If he thinks it will strengthen he might want to agent a deal and set the credit spread later. This must be considered in conjunction with overall spread trends in the market and underlying interest rate trends as well.

ZI, Goldman Sachs  We are real estate people, so we are happy to pay for Jim or Cyril to take that risk.

CC, Deutsche Bank  We've slightly moved the conversation. The topic until now has been about real estate lending and the fact that an increasingly significant component of that real estate lending ends up in a securitized market. And as we've just heard, borrowers can obtain liquidity in the loan market very quickly in the context of a competitive auction. However we are now talking about agented securitization, and there is a very important distinction between the costs on a real estate loan which is later securitized, where you pay arrangement fees but no securitization fees, and an agented transaction. The timing element changes as well: it doesn't take two to six months to structure and fund a loan.

RG, Morgan Stanley  For many borrowers, ourselves included, on an acquisition, the reality is we just don't have the luxury in most cases to defer the purchase price sufficiently to do an agented securitization. So despite the higher cost of a principal transaction, we often just have to build that into our acquisition price and move.

JB, Lehman  I think most agency transactions are well structured, but they are not for everyone given the opportunity cost. For small transactions agency transactions are not a good use of the CFO's time. He can get a loan in three weeks instead of spending six months going on a roadshow. Also, in the agency business there are banks so desperate to do a deal that they will persuade borrowers to do deals that will not trade or that are not properly structured and I think that hurts the market.

AR, NM Rothschild  I'm going to play devil's advocate here. Take, say, a £200 million loan; that will be done either through a conduit multi-borrower deal – so it will end up getting securitized anyway – or a stand-alone. The banks will make more money if it goes through a conduit deal, but if the borrower takes market risk themselves and issues at a set price, then hasn't the borrower got a better deal going direct to market if they can manage the process?

CC, Deutsche Bank  No, I completely disagree.

JB, Lehman  In the conduit deal we will spread the securitization expenses over a bigger pool.

CC, Deutsche Bank  If I'm a shareholder in a small UK plc that owns £300 million of assets that's looking to borrow £100 million on an agented basis, do I want to see my CFO and my CEO and six of their brightest people working for two, three, or four months on an agented securitization to save 25 basis points? Absolutely not. I want them working on new acquisitions and the underlying business.

PC, British Land  The true advantage of doing an agency deal – although you can do it with a conduit deal to some extent – is the individual flexibility you build into your deal in terms of what you want to do with an individual portfolio of properties. Putting your assets into a conduit, which has a standard set of terms and conditions, even though it might achieve best execution in terms of price, won't necessarily achieve best execution in terms of the flexibility and lack of restrictions.

AR, NM Rothschild  The key issue is the role of the bank in reducing the opportunity costs that Cyril is talking about, and this has changed a lot. The banks and the clients recognize the opportunity cost and banks have to take over much of the project management. I believe that in a proper client-bank relationship it is the bank's responsibility to maximize efficiency and minimize the client's time spent on the transaction. Going back to a point earlier, clients want to spend their time buying property. It does require additional resource investment from the arranger's side, but it results in a far happier client.

JB, Lehman  We recently did a loan for a very sophisticated borrower. We committed to the deal in less than a week and closed in three weeks. Is that a loan they could have securitized? Theoretically yes; what's the most important thing to the borrower, 20 basis points or making sure the deal closes?

ZI, Goldman Sachs  When we're looking at a new acquisition we would never even think about an agented deal; the only time we'd think about an agented deal is when it's a refinancing. You have the luxury of time and you already own the risk. The other structure worth mentioning is market flex. If the bonds price worse than where you expect, then your downside is capped as a borrower, but at the same time you get a share of any upside if they price better.

CC, Deutsche Bank  There are a number of permutations between the fully underwritten and the fully agented deals; you have varying degrees of costs, risk transfer, risk allocation, resource allocation and resource requirements. The point that I think Jim makes is that an agented deal is used probably more often and less efficiently that it might otherwise be, and perhaps the fully underwritten deal is taken as an easy choice, but not necessarily used in the most efficient way either, and there are various degrees of choice between a) and b).

AC, Clifford Chance  I think it depends on complexity. If the deal is complex, then agency might be the route; if not conduits are fine.

SO, Brixton, left the discussion at this point.

The future of the CMBS market

SB, Euromoney  So what do you all see developing in this market in the next two to three years?

JB, Lehman  I think we will see continued growth of securitization. Most loans that have a cashflow characteristic will end up in the securitization market. One thing I hope to see is being able to get deals rated more quickly. Although this has got significantly better, I'd like it to be possible to get a deal rated in two months.

We are also seeing more acceptance of structures which don't fit the classic first mortgage/clean SPV structure which is the bedrock of our business. This will grow and be helpful in terms of giving additional flexibility to borrowers. I also think that we can look forward to lower credit support levels on the continent, which will make securitization more attractive to a borrower versus a real estate loan provided by a securitized lender. A lot of the big, international lenders – particularly some of the German lenders – will see their bread-and-butter business drop. Any asset that can be securitized over the next three years will most likely be lent on by a lender that exits via the securitization market because banks that cannot fund themselves through the securitization market won't win many loans, because they won't be competitive.

ZI, Goldman Sachs  For me the question is sustainability of spreads as that is a key determinant of the viability of the securitization market. Spreads are at amazingly low levels – we've invested a lot in mezzanine transactions, and we've had paper quoted at 600 over 12 months ago trading 200 basis points tighter today – and I wonder if this environment is here to stay.

JB, Lehman  I think it is – the whole credit curve has tightened and flattened and that looks here to stay. And in your particular case – mezzanine – something else has changed: there are a lot more people investing in mezzanine and B-notes now than there were two years ago because of people like myself and Cyril saying that there is a credit relative value and spread relative value in buying real estate risk at the mezzanine level in Europe versus the US.

PC, British Land  And also spreads in real-estate linked securities have reacted to the yield compression on property itself. The narrowing of property yields has meant you've got to be able to finance more efficiently in order to do deals, and the health or otherwise of the UK and European property markets is going to be one of the things that drives the health of the CMBS market.

JB, Lehman  I think that's right and I think both of them have been big beneficiaries of the leverage in the system, as well as the attractiveness of cashflow assets versus the stock market. So if the UK property market goes into a slump, you will see spreads widen, because the perceived risk will be much wider. But I do think that that is only part of it. Regardless of property fundamentals, the asset class is more accepted and there's a wall of money that's chasing both debt, spread product and property assets today.

ZI, Goldman Sachs  That was my question: what could affect the wall of money?

JB, Lehman  One thing would be transactions going wrong in ways that are not explainable in terms of defaults. It is a relatively young market and we have had a great property market to build it in.

ZI, Goldman Sachs  One thing that would make life easier for us is more standardized loan documentation. If we do a deal in Italy versus Germany you're starting from scratch and you have to renegotiate the loan document all over again. Better legislation would also certainly help in the market.

ACh, Euromoney  Can lawyers have an input there?

AC, Clifford Chance  Yes, and I think it is slowly happening, but there can be resistance from each country to accommodate initially, because this kind of thing gets viewed as a sovereign issue. However, the CMBS market, probably more than anything else, is Anglo-Saxonizing the European property finance markets and that's certainly introducing more competition into European finance. AR, NM Rothschild  I think that we're at an interesting juncture in the market. I think Europe is certainly an issue and we've seen in Germany that there's been a loosening of objections to standardization of documentation. I think that that process will be speeded up as issuance volumes increase. An interesting question is how accessible this market will become to a wider set of borrowers. How will pricing benefits be passed across to borrowers? Who will use the market? Can we create better hybrid products between agency and conduit? How are the up front costs defrayed? Are they through the spread or can the borrower pay them? I also think there has to be rationalization in the conduit business. Too many individual conduits have been set up by too varied a quality of people, and there is just not enough business for people to be issuing their own £100-odd-million bond for purposes of ego. So I think there will be increasing pooling of assets by banks and a core of banks will emerge as the distributors and originators.

Once that happens liquidity will improve, the frequency of issuance will improve, and the banks will carry much less spread risk, much less origination risk, much less cost risk, and benefit from economics on being able to participate in the tranching benefits. When that starts happening you'll see many more borrowers able to access the market.

RG, Morgan Stanley  A couple of things are happening that we look forward to seeing accelerate. First, the B-piece market here is developing fast – it is getting stronger every day. The investment banks and other institutions are staffing up in this rea, bringing in good people, who will be a major force in the marke. That will make a substantial difference, we believe, in the trading of the B-notes and those kinds of developments will add to the weight of other strong buyers like Hypo and EuroHypo. That relates to another development that we look forward to seeing more of, which is secondary trading in the market.

The reason that these developments are helpful for us, even though we're just a borrower, is that we try to differentiate ourselves in our long-term performance with a measurable high-quality credit history, and we hope in the long run that should give a different cost of capital for some owners of property versus others. And if the rating agencies would staff up more, it would certainly speed their turnaround times and generally help us execute deals more quickly and help our friends on the lending side be that much more certain about deal outcome, which helps us remain at our most competitive in acquisitions.

I'd also like to talk about the changing cost of capital. We've seen, for example in Italy, a very dramatic shift in the cost of capital. Previously Italian lenders had a lock on their domestic market. Being able to bring in capital from outside Italy drove down the pricing of that capital significantly and helped accelerate the number of transactions that closed there. The government seemed to vacillate about whether it needed to enact protective legislation for its local banks, or whether such legislation would damage the Italian commercial property markets and ultimately negatively impact their economy and citizens as a result. They came out on the side of cheaper capital, seeing that that is good for Italy and that protecting Italian institutions would be damaging to Italy's long-term financial health. And we see that same mindset of capital freedom generally taking deeper root across all of Europe.

Dramatic change

JB, Lehman  I think what's really impressive with this market is how it's changed over the past five years. It started as a niche product for certain borrowers and has moved into the mainstream with really significant adoption of the technique.

One thing we haven't touched on is how much the investor base has grown. I can only imagine the roadshow in 1999 on the first Broadgate deal, and I know our first Windermere deal (2001), the first Italian deal (2002), the first Swiss deal (2002), the first Swedish deal (2003) – all these were massive events; we made tremendous efforts educating a lot of investors. Today, investors are now much more sophisticated and educated.

What's important now is to make sure that as the market matures we solve some of the things that are not as strong: we need better disclosure, better information about the bonds, more ability to trade them and we need to make them more flexible. But the market is here to stay and it is going to grow.

SB, Euromoney  OK, great, I think we can wrap up there. Thank you all very much.