iStar Inc (STAR) 2007 Q3 法說會逐字稿

Welcome to the iStar Financial third quarter 2007 earnings conference call. (OPERATOR INSTRUCTIONS). As a reminder, today's call is being recorded. At this time I would like to turn the conference over to the Vice President of Investor Relations, Mr. Andy Backman.

Thank you. Good morning, everyone. Thank you for joining us today to review iStar Financial's third quarter 2007 earnings report. With me today are Jay Sugarman, Chairman and Chief Executive Officer, Jay Nydick, our President, Tim O'Connor, Chief Operating Officer, and Katy Rice, our Chief Financial Officer. This morning's call is being Webcast on our Web site at istarfinancial.com in the investor relations section. There will be a replay of the call beginning at 1:30 PM Eastern Time today. The dial-in for the replay is 1-800-475-6701 with a conference confirmation code of 891239.

Before I turn the call over to Jay, I need to remind everyone that statements in this earnings call which are not historical facts may be deemed forward-looking statements. Factors that could cause actual results to differ materially from iStar Financial's expectations are detailed in our SEC reports. Now I'd like to turn the call over to iStar Financial's Chairman and CEO, Jay Sugarman.

Thanks, Andy. Thanks, all of you, for being with us this morning. Since this is the first earnings call that includes the discounted purchase of the Fremont commercial loan portfolio, we're going to spend a good amount of time drilling down into the numbers for you, in addition to reviewing our regular quarterly results. Given current market conditions, we think detailed knowledge will go a long way toward helping put to bed the uncertainty and potential concerns that appear to be impacting our company and our valuation. Let's go to the third-quarter results first, and then start the deep dive afterwards.

The third quarter was a busy one for our company, with solid earnings and an all-out effort to integrate the Fremont portfolio and its employees. We made good progress on all fronts, and have begun reworking the portfolio to shape it to current market conditions. Here are the highlights from the third quarter.

First off, earnings were a very solid $1.07 per share. Katy is going to walk you through how the Fremont portfolio and its discounted purchase price came onto our books from an accounting perspective, and then later I want to focus with you on how we look at the assets from a pure economic and return on equity perspective.

On the origination side, we closed over 40 separate transactions and, excluding Fremont, funded over 1.3 billion on new and previously committed transactions. Now, while pricing on legacy transactions in the real estate pipeline were not very interesting, we did see very good opportunities in the corporate market, as the pricing dislocations in that market were much swifter and immediate than in the real estate sector. With the corporate market inching back towards stabilization, we think the best opportunities in the next several quarters will be back in the real estate world, where things are finally beginning to reprice.

With regards to return on equity, we had a strong quarter, with return on equity finishing over 21% as we move closer to our long-term capital model leverage targets.

On credit, we added the Fremont watch list and NPL assets to our own small number of problem assets, and are now grinding away on those assets to get them to resolution. After our recent quarterly asset-by-asset review of all 650 assets in the portfolio, I take comfort that the vast majority are well-protected, and still feel very comfortable saying that potential embedded gains in our portfolio are well in excess of any potential losses. I'll share some detail on that point when we go to the deep-dive analysis.

Let me turn it over to Katy now, and then come back and share some specific portfolio statistics with you.

Thanks, Jay. Good morning, everyone. As you know, this is the first quarter that we're reporting the impact of the Fremont acquisition in our numbers. There are a lot of moving pieces and many different ways to look at the transaction. We have attempted to provide you with a clear view of both the initial transaction, and give you an update on where we stand today versus our original expectations. Before getting into the specifics of Fremont, let me walk you through the third-quarter results.

As Jay mentioned, we had a strong quarter. Our adjusted earnings came in at a record 136 million, or $1.07 per diluted common share. Our net investment income was 221 million, up over 100% from the third quarter of 2006. The increase was primarily due to the addition of the Fremont portfolio.

Excluding the Fremont transaction, new commitments for the third quarter totaled 1.8 billion in 42 separate transactions, 988 million of which was funded during the quarter. We also funded 397 million of pre-existing commitments, and received repayments of 401 million, for a total of 983 million of net asset growth for the quarter. Approximately 79% of our third-quarter commitments were first mortgage or other senior debt commitments, and 13% were new CTL originations.

Our adjusted return on equity came in at a very strong 21.8% this quarter. Our net finance margin for the quarter was 5.25%. This quarter's net finance margin was positively impacted by the Fremont transaction. Specifically, $60 million of amortization related to the purchase discount of the portfolio. Excluding the effect of the purchase discount amortization, our net finance margin for the quarter was 3.53%, versus 3.22% from the prior quarter.

With respect to our credit statistics for the third quarter, interest coverage was 1.8 times, compared to 2 times last quarter. Fixed charge coverage was 1.7 times, compared to 1.9 times last quarter. The decline was primarily due to increased interest and G&A expense, some of which were onetime expenses associated with the Fremont transaction, as well as the ongoing cost associated with the additional personnel and our eight new offices across the country.

Our leverage, defined as debt to equity plus accumulated depreciation, depletion and loan loss reserves, was 3.3 times at the end of the quarter. The increase in leverage was primarily due to the acquisition of Fremont and the 100% first mortgage composition of that portfolio, and it is within our current mid-three times target leverage range.

Let's move on to credit quality. As you know, risk management is a critical part of our business strategy. We have nearly 140 people at the firm devoted to tracking, managing and understanding the risk associated with our portfolio. This quarter we undertook the same rigorous review and risk rating assessment for the Fremont portfolio as we do with our own portfolio every quarter. Based upon this review of all of the assets, including Fremont, 11.5% of our structured finance assets ranked four or higher. Of the assets that ranked four or higher, 92% were on our watch list or on NPL status.

With respect to our CTL portfolio, we saw a slight improvement in the credit metrics this quarter. At the end of the third quarter, the portfolio included 407 properties with 41 million square feet of space leased to 123 different credits. The portfolio is 95.6% leased, with a weighted average lease term of 11.2 years. 2008 lease expirations are minimal and represent just 0.6% of annualized revenue.

As we expected, and as we outlined on the July earnings conference call following the close of the Fremont transaction, our NPL and watch list assets grew significantly with the inclusion of the Fremont assets. I'll break the two portfolios out for you, and then give you a little more color on the trends we see in the portfolio.

With respect to the iStar portfolio, the NPL list remains essentially unchanged from last quarter. At the end of the third quarter we had seven assets on our NPL list with a total book value of $209 million. As we discussed last quarter, most of these loans were higher risk-adjusted transactions that were underwritten with higher return expectations.

With respect to our watch list this quarter, we added two loans totaling $155 million. We now have six loans, which have a total book value of $315 million. The largest of these is a corporate real estate loan to a company backed by a major private equity sponsor.

Now let's turn to the Fremont portfolio. At the end of the third quarter we had 22 Fremont assets on NPL status totaling $640 million in book value, up from 16 assets totaling $406 million in book value at the end of the second quarter. One thing to keep in mind when you look at the Fremont numbers is that they are grossed up for the A-participation.

We expected the NPL list to increase this quarter, and expect it to increase over the next couple quarters, although more modestly. Loans on the NPL list this quarter range in size from $7 million to $75 million.

At the end of the third quarter, Fremont watch list assets included 22 first mortgage loans totaling $781 million of book value, versus 22 loans with a total of $844 million of book value in the second quarter. The loans range in size from $10 million to $100 million. Our asset management team is making progress in resolving issues on a number of these loans; however, we expect the process in most cases will take several quarters to reach resolution. While the amount of Fremont NPL and watch list assets is slightly larger than our original underwriting, some assets in the portfolio are performing better than expected. So the overall results thus far are generally consistent with our underwriting.

Now let's switch gears, and I want to walk you through some of the changes from the way we originally expected to account for the reserve associated with the Fremont transaction. Originally we anticipated transferring Fremont's then-existing loan loss reserve of approximately $210 million into our reserves at closing. Subsequent to our closing with Fremont, they decided to go back and impair their loan book by the amount of the discount we negotiated. Once they impaired the loans, the loan loss provision was eliminated, leaving no reserve balance to transfer. In lieu of booking the actual loan loss reserve at closing, we booked each individual loan at fair value, including specific impairments, and will now amortize the total discount of approximately $280 million over the life of each loan into interest income to the extent we believe we'll collect it. 18 of the loans have specific impairments, which totaled $57 million. This amount is included in the total discount of $280 million, but will only be accreted into earnings at such time as we believe it will actually be collected. For example, this quarter, approximately $60 million of discount associated with the portfolio was accreted through interest (inaudible) income.

The net effect of this change to the accounting will not generate a significantly different economic result; however, I wanted to make sure that you understand why you do not see a much larger total provision amount at the outset. Under this accounting methodology, we will take reserves for the Fremont portfolio on a quarterly basis, based on our review of the portfolio.

This quarter we added $62 million to our reserve, bringing our total loan loss provision to 124 million, or 1.16 of total loan assets and 15% of total NPL assets. The increase in reserves was the result of our regular quarterly risk rating review process, as well as the addition of the Fremont portfolio. We expect to continue to build reserves based on our quarterly assessment of the risk we believe exists in the combined portfolio.

Before I turn to our guidance expectations, I'd like to review our funding needs and our capital markets activities. One of the current misconceptions we've heard from a number of investors is the size of our funding needs related to the Fremont transaction. Let me walk you through exactly what our funding needs are.

When we announced the Fremont transaction in July, we had unfunded commitments related to the portfolio of approximately $3.7 billion. By the end of the third quarter, these unfunded commitments had dropped to $3 billion, due to fund-up activity in the quarter. In addition, we believe approximately $400 million of these commitments will never fund. Many of these commitments are construction loans for projects which will not move forward or have not met the pre-sale milestones required in our loan documents.

In addition, we've identified approximately $200 million of unfunded commitments in our core portfolio that we believe will not be funded. We expect to identify other early-stage projects that will not move forward, further reducing our unfunded commitments.

So, for the remainder of 2007 and through the end of 2008, our combined unfunded commitments total approximately $4.8 billion. During the same period, our asset maturities total approximately $5.4 billion. So this part of our portfolio is effectively self-funding, and in fact will generate some additional liquidity.

One of the things that people often lose sight of is that the weighted average maturity of our structured finance assets is just over three years. In fact, over the past couple of years, we've intentionally originated shorter-term loans that we expected to repay relatively quickly, giving us the ability to redeploy capital at higher spreads as the market turns.

In addition to the repayment and funding activity, we also have approximately $2 billion of debt maturities in 2008, inclusive of the interim credit facility used to finance the Fremont transaction. We currently have approximately $1.5 billion available on our revolving credit facility, so our net funding requirements are actually quite small.

It is also worth remembering that we have over $15 billion of unencumbered assets, many of which are readily financeable even in today's market. And given the size and diversity of our asset base, we have tremendous funding flexibility, and we expect to raise capital in the markets that offer the most attractive cost of funds.

Finally, let me wrap it up with a discussion of our earnings guidance and our dividend increase. As we said when we announced the Fremont transaction, we expected the acquisition to be accretive this year. Based upon our current view through year-end, we expect 2007 diluted EPS of $4 to $4.10, and diluted GAAP earnings per share of $2.80 to $2.90. These ranges are based on net asset growth in 2007 of approximately $5 billion, including Fremont.

Given the disruptions we've seen in the markets as a result of the credit crunch, it's more difficult to predict the overall environment in which we will be operating in 2008. What we have done with respect to 2008 guidance was to provide a framework with which to think about our 2008 earnings. We will, however, update you over the next quarter or two as things unfold in the market.

To the extent the capital markets remain inhospitable, we will slow our growth in 2008. In order to help frame the low-end of an earnings range, we took a look at what is admittedly an extreme example, and that is zero net asset growth in 2008. This scenario would require minimal funding; however, we would still achieve adjusted earnings of approximately $4 per diluted common share and have good coverage for our dividend. However, as Jay mentioned, this is the market environment that we are built to take advantage of, and we're been beginning to see attractively priced transactions that are accretive, even at a higher cost of funds.

So, looking at a more normalized scenario, if we assume 3 to $3.5 billion of net asset growth in 2008, with the growth weighted in the latter part of the year, we would expect an EPS of approximately $4.20. So, with that as a backdrop, we're giving you an EPS guidance of $4 to $4.20 for 2008, and we will update you as markets develop and we have better visibility.

Last, but certainly not least, let's talk about the dividend. As you know, we typically review our annual dividend rate in the first quarter of the year. However, as you saw in the press release this morning, we announced that the Board has approved a 5% increase in our regular quarterly dividend, effective immediately, with the next quarterly dividend payable in the fourth quarter, and through the end of 2008. That brings the annualized dividend rate to $3.48 from $3.30 per share. Even with this increase, we expect to declare a special dividend in the range of $0.15 to $0.30 per share before the end of the year. This special dividend will help ensure that we pay out 100% of our taxable income this year. The increase in taxable income this year is primarily related to the acquisition of Fremont.

So with that, let me turn it back to Jay.

Thanks, Katy. So, two important points Katy made that I just want to reemphasize.

First, even with zero net asset growth next year and an increased cost of funds, we're still projecting close to $4 per share. That's, obviously, not what we think will happen, but it gives us quite a bit of comfort on the dividend and quite a bit of control over how and when we play next year, which really leads to the second point.

If asset spreads and liability spreads don't make sense, we will not be issuing liabilities or raising unsecured capital until they do. This company has tremendous flexibility on the funding side and in different markets, and we're going to be very focused on raising capital that can be accretive to earnings.

Let's move to some of the drill-down analysis we use internally when thinking about credit and returns on the portfolio. First, let me take you through our overall principal risk ratings.

As you all know from prior calls, we rate all loan and CTL assets on a scale of 1 to 5 in increments of 0.5 point. So investments rated 1.0, 1.5 or 2.0 are high-quality, good-things-are-happening assets. Investments rated 4.0, 4.5 or 5.0 are higher principal risk, not-such-good-things-are-happening assets. Here's how the ratings came out at the end of the third quarter.

For the iStar loan portfolio, the 1's, 1.5's and 2's were $1.855 billion, while the 4's, 4.5's and 5's were $910 million. For our CTL portfolio the good stuff was $1.800 billion, while the not so good stuff was $550 million. And for the Fremont loan portfolio, the 1, 1.5's and 2's were $385 million, while the 4, 4.5's and 5's were $990 million.

So, with respect to overall value in the book, we continue to believe there are far more assets with unreported potential gains in the portfolio than unreported potential losses. Remember also, we are reserving upfront against most of the 4's and 5's ahead of the losses, but not booking anything for the good stuff until it's actually realized, so there's also positive bias there.

A couple of other points on credit. First, the top 10 assets now make up only 12% of total assets, down from 23% just over a year ago, so the granularity of the portfolio has increased dramatically. That diversity is a major credit positive for bondholders.

Second, a certain amount of asset stress will likely create some upside opportunities to offset some of the obvious downside. One of the best short-term investments we will make in the second half came about due to a loan extension of a large asset under some stress. While the asset is highly levered with multiple layers of subordinate capital, our first mortgage is at a 40% loan-to-cost level, with several hundred million dollars of subordinated capital behind it. And we're going to earn a very attractive return while that loan remains outstanding.

Third, while it seems clear no one really knows how to value sub-prime residuals, or how to work out little slivers of complicated financial products, large commercial real estate first mortgages are underwriteable and can be worked through with some level of confidence. We are taking reserves as appropriate for that portion of the book that can be reserved against, and continue to take comfort that these reserves are based on real facts and reasoned judgment rather than statistical models.

Lastly, while we don't know exactly how the world will look over the next few years, I think we can say categorically our bonds are absolutely well-protected under any scenario. Current iStar bond spreads seem to be an extremely compelling investment, and one of the best risk/reward opportunities I've seen in the high-grade market in a long time.

Now let's turn to returns on the Fremont assets. The number-one question we get asked is if our underwritten returns will be dragged down by unexpected losses. Let's focus on this question for a minute.

Let me start by laying out the assumptions underlying our return model. We start with the original purchase of $6.2 billion of first mortgages, with principal funded 70% by the Fremont A note and 30% by iStar. A total of approximately $3 billion of add-on investments will be made over the next 24 months, funded 100% by iStar. iStar will leverage its first mortgage positions in accordance with this capital model at approximately 4 to 1. The portfolio will start with a total of approximately $700 million NPLs that earn nothing until [the later of] the third quarter of 2008 or six months after maturity. And recoveries on those NPLs will only be 62% of face.

Based on those assumptions, the return on equity on the Fremont assets will be 24.5% if our cost of funds is LIBOR plus 50, which was the cost of the initial bridge and our existing unsecured credit facility. It will be 22% if our blended cost of funds ends up being LIBOR plus 100. It will be 19.5% if our blended cost of funds is approximately LIBOR plus 150. And lastly, we run a case where our cost of funds starts at LIBOR plus 50 and rises to LIBOR plus 200 over the next three quarters. That case also ends with a return on equity of 19.5%. So, the base case return on equity range is effectively 20 to 25%, depending on our actual cost of funds.

Those returns are predicated on our discounted purchase price, which was $280 million below face, which works out to around 40% of the face on the assumed 700 million of first mortgage NPLs. Let me just put that in perspective for you. Losing 40 points on a loan that was originally underwritten at 80% loan to cost implies you are recovering only $0.50 on the $1 of the assets' original cost. So, if you funded $100 million loan to build something, and you lent 80%, or $80 million, we are using a recovery value of 60% of the loan, or $48 million. Obviously, $48 million is less than 50% of the original $100 million it cost to build the asset.

Our own experience is losses of that magnitude are very infrequent and will not represent the outcome of all or even a majority of the assets we are working through. As you might guess, we've run many different combinations of NPL amounts and principal loss scenarios, and we think it's probably most interesting to assume a scenario where the portfolio ends up taking about a 20% discount to face on about $1.4 billion of problem assets. That case still generates a 20% return on equity.

So, based on that analysis, plus the anecdotal evidence we have to date about workout values on similar assets, we continue to believe that our base case underwriting of $280 million in discounts to face will prove to provide us a sufficient cushion to work through portfolio issues and make that 20% ROE return. We'll walk you through some of those anecdotes in the next quarterly review when we have a more statistically meaningful sample.

Let me now turn to our current investment strategy, and then we can open it up for questions.

In terms of investing, we think we're moving into a very attractive period for many parts of the core business, with a return to pre-CDO pricing, and the opportunity to receive returns based not only on underwriting credit risk, but also for just providing liquidity. We've said repeatedly over the last two years that the market was too bullish and was understating credit risk and charging virtually nothing for liquidity. With the hard credit crunch taking place, the areas we will try to emphasize are likely to be where liquidity and underwriting will be paid handsomely and where call protection can be achieved. We've spent a lot of time focusing the team on finding opportunities that represent the best risk-adjusted returns available right now, wherever they may be. It's not clear to us that the pricing matrix in real estate has moved far enough to hit the price and structures we are currently quoting. But for now we're being patient and trying to consistently show the market where we think risk and liquidity should be priced. A few deals actually have closed at these new levels, more have not, and many are simply waiting to see what happens.

In addition to the core businesses, we're also continuing to look for ways to expand our platforms and add to our growing network of knowledge across the real estate, capital markets, and corporate credit arenas. Jay Nydick has continued to see interest from outside sources to invest in our extended platforms, and we expect to be positioned to capitalize on those opportunities by leveraging our own capital with third-party capital wherever appropriate.

And one last note. We will be adding a new board member in the next several quarters, as Carter McClelland has recently informed us he has accepted a new position as head of a start-up investment firm on the West Coast that will, unfortunately, prevent him from continuing to serve on our board. We wish him well in this latest endeavor.

With that, operator, I think we can go ahead and open it up for questions.

(OPERATOR INSTRUCTIONS). Don Fandetti, Citigroup.

Thanks for all the color on Fremont; that's very helpful. I wanted to ask you broadly, in terms of your outlook on commercial real estate over the next few years, there's a very bearish view being -- developing in the marketplace. I wanted to just get your thoughts on that as it pertains to your legacy business.

I guess we're watching a lot of the early indicia that suggests people are getting very bearish as well. I guess right now I'd break it down into the business side of commercial real estate still seems to be doing very well; industrial; hospitality; those things that are tied to the business cycle are doing pretty well; those things that are tied to the consumer, probably less well. Certainly, the housing market and retail are two of the areas that we're going to have to watch closely.

My own view over the long periods of time in real estate is replacement cost and location still govern most of the things we look at. If you're in good real estate, well-located, well-built, below replacement cost, ultimately, things tend to find that level and then move beyond it. So, I think our lending discipline still is stable over placement cost and high-quality assets and good locations. And I think that's probably true for all commercial real estate is -- the stuff that has a reason for being, that has a proper manager, that is well-supported; we still feel pretty comfortable that's a very nice place to invest.

I think when we look back over the last couple of cycles, fundamentals in real estate, supply and demand in real estate in most areas are still pretty good. Not great. I'd say office is kind of in the middle of the pack; some markets feel good, some don't. But overall, we don't see the fundamental stresses, nor did we see nearly the level of craziness that you saw in the residential markets. So we think some of the anticipation that the residential scenario could play out in commercial is just not right.

Lastly, on Fremont, what do you think investors -- what metrics should investors look at over the next six to nine months to judge the health of the Fremont condo portfolio? Could you provide some color there?

As we did today, I think, we're going to give you quite a bit of detail on anecdotes about things that are happening, both on the NPL and watch list, but also in some of the other areas. Again, I think, well-located projects with strong, well-capitalized sponsors -- they're going to stick with it. That's historically what we've seen. In many of the cases we have large institutional mezzanine lenders also beneath our first mortgages, and they, too, I think, have a much longer-term view than what's happening today or tomorrow. Again, at $0.50, $0.60, $0.70 on the $1 of replacement cost, with steel, glass, concrete all still rising in cost, I don't think anybody is going to walk away lightly from a brand-new building in a well-located position. We've done a lot of work on this portfolio, both before we bought it and since we bought it, and we'll be able to give you quite a bit of explicit detail over the next couple quarters.

Douglas Harter, Credit Suisse.

I was wondering if you could just remind us how the risk sharing between the A piece and the B piece on Fremont works.

We show the entire amount of a first mortgage in our NPL or watch list. They receive $0.70 of every $1 of principal returned. We receive 30%. But as you know, the $3 billion of future fundings also go into that repayment basket. So they've got additional credit support from there as well. So we think it's appropriate to show the whole first mortgage as a risk asset of iStar, rather than only showing the 30% that we actually own today.

So if there's a loss, does Fremont share in that loss, or do you absorb that entire loss?

Effectively, they get $0.70, we get $0.30. But they can draw that $0.70 not just from the original assets, but also from the forward funding assets. So we believe all the losses will effectively be iStar's.

So the gross NPL number is probably the better number to look at?

That's why we show that number.

Susan Berliner, Bear Stearns.

Katy, I was wondering if you could -- that was helpful going through your funding needs. But, with the bridge facility, as well as some bonds maturing in March, I was wondering if you could tell us exactly how your -- or what options you have to fund going forward, and what your timing is.

As we mentioned, the funding and repayments actually generate positive liquidity in the portfolio. So we don't have to worry about that. But with respect to maturities, we do have credit capacity on our lines of credit. And as I mentioned, we do have $15 billion of unencumbered assets. I think we're at an all-time high right now, in terms of levels of unencumbered assets. That provides tremendous flexibility for our funding. And I think we'll be looking -- to the extent the equity and unsecured markets remain somewhat difficult and dislocated, we will be looking to other areas within the portfolio that we can generate funding that have a much more attractive cost of funds for us.

I guess have you gone over with the rating agencies what your funding plans are? And if you could just tell us what their comfort is with your funding as well as your leverage.

We have been in discussions with the agencies, and we've been keeping them updated. Since they affirmed our ratings in July, we're not aware of any major concerns related to Fremont or any other aspect of our business. They are, of course, focused on capital structure, but they also understand the significant flexibility we have with respect to our unencumbered asset-base. And I think they understand that we'll need to access capital opportunistically, given the high degree of volatility that we're seeing in the markets right now.

David Fick, Stifel Nicolaus.

I'd like to just walk back through guidance a little bit. You did a good job of talking about sort of worst-case. But then the high-end of guidance is only $0.20 higher. And Jay, I think, your words were this isn't what you think is going to happen. Then why is the top-end only 1 to 4% growth, given the positive profile that you have on recycling your capital, and the thought that you'll do more than $3 billion at the high-end of net increment asset adds?

I think what we really wanted to do was lay out a conservative scenario. I think right now, given the lack of really clear visibility in the next couple quarters with respect to both asset spreads and liability spreads, we really tried to create a kind of earnings framework with a conservative range on the top-end. I think really what will probably realistically happen is relatively slower growth over the next couple of quarters. And then as things evolve, hopefully, growth will pick up from there.

Here's the math issue for us. If you put assets on in the third and fourth quarter next year, they almost don't impact '08 earnings. So it really comes down to do our credit spreads and asset spreads make sense in the first and second quarter to put on the volume early? Or, as Katie's case really outlined, putting a lot at the end of the year is good for '09; it doesn't do much for '08. I think for me personally, I think, the asset spreads are going to be there. But as I said, some people are hitting our bids; some people are not. If this market were to tighten up as quickly as we saw the corporate market sort of realign itself, we're not going to get hit on a lot of those structures in pricing. If it doesn't, we'll get hit a lot more often, we'll put on a lot more assets early in the year, and that would be my own internal upside case. It's not prudent right now to tell you that's what's going to happen because we're not seeing it happen yet.

So the real message in this guidance is a lot less pipeline visibility with back-loading on the assumption that the market is going to rationalize?

And that's a function both of credit spreads, which we talked about. We're scratching our heads a little bit on those. Certainly if those credit spreads were to come back in quickly, there's a lot of stuff to do out there right now. But we're not going to force it. We're not going to put on assets and liabilities that don't make sense for us right now.

A related question is what assumptions are you making on G&A and loss reserves in that guidance?

With respect to G&A, we have seen an uptick, obviously, with respect to Fremont. And the integration has gone well to date; we've seen a lot of synergies from the business. Obviously, we expect, as part of our ongoing efforts to manage G&A, we'll continue to look to ways to reduce those expenses. Our assumptions in 2008 reflect the full year of Fremont G&A, as well as an increase in expense associated with our new lease that we signed for our New York headquarters across the street from our current location. So we're assuming an uptick in G&A of about $30 million from '07 to '08.

With respect to reserves, obviously, we look at this quarterly with what we actually see happening vis-a-vis the quarterly risk rating process. But from a modeling perspective, we assume additional reserves are taken sort of heavily weighted in the early quarters in the next five quarters--heavily weighted in the beginning, and then tapering out a little bit towards the end of '08. And I think those total somewhere in the neighborhood of 100 to 120 million.

Michael Dimler, UBS.

Good morning. I wonder if you could talk, at least generically -- I don't think you can get down to specifics -- but, the characteristics of the rise in delinquencies and nonperformers in your core portfolio, ex-Fremont.

It's not much different from last quarter. It's the same assets, same story. There's very high coupon first mortgages and second mortgages in there, anywhere from LIBOR plus 700 on the low-end to 30% on the high-end. Some of them were the seminal investments in what we thought could be interesting platform opportunities. We're learning the hard way those returns are hard earned. And so, as we said on our last call, I think those are far more one-off in nature, and there's just not a repeated pattern throughout the portfolio.

I think when we look at the Fremont side of the equation, there is some repeating pattern, although we think there's great diversification inside that book when you include it in our overall portfolio. So again, when we look at the individual issues in some local submarkets, west coast of south Florida, those are small relative to overall booked. But they are two, three, four assets in a market all having the same problem. Those are more systemic to that local market, and those are the ones we're working very hard on. Some of them we're getting out of, and getting out of at decent prices. And others we're going to have to work very hard just to get the book back that we thought. But they tend to be small in relation to the overall portfolio.

You mentioned valuations a little bit ago. How do you factor cap rate normalization into your return analysis, at least on -- how would that impact existing assets if you see cap rates on real estate revert to more historical levels?

Again, if you've been following us for a while, we have historically tracked the internal value we assign to assets versus what they actually either sell at or get refinanced at. And historically -- and I'm looking at Tim O'Connor -- we have a list that, I think, says on average we understate the value by 28%. So if we think it's worth 100 million, and the borrower goes and sells it, he's getting 128 million, or he's refinancing it based on a $128 million valuation.

So, one thing I would just point out is we don't mark our valuations to the last trade in the market; we mark it to what we at iStar would pay for that asset in a normal market already. So we don't have four caps and five caps and three caps. We think things trade historically over long periods of time close to replacement cost in what I would say is probably 100 basis points higher than what you saw in the first six months of this year, last six months of last year. So I don't think there's a lot of revaluation going on in our portfolio. The cushion -- I'll bet you that 28% is now down to something materially lower. One of the statistics we always look at is our underlying CTL book, which is comprised mostly of publicly traded and high-quality credit. It is over a nine cap. And I can tell you, based on stuff we see trading right now, that is several hundred basis points wide to where those would trade in the open market. So it's not a five cap world; it's probably up 100 basis points, 150 basis points for high-quality assets, and maybe even a little more for low-quality assets. But at the same time, I don't think we ever moved our valuations up. So all that's really happened is the cushion has been eroded closer to the value we thought was appropriate in the first place.

[So you kind of] value through the cycle is what you're saying there?

Yes.

Mark Streeter, JPMorgan.

Jay, I'm wondering, just looking back in hindsight, I know it's difficult to do, but with the Fremont acquisition, given the way the market has reacted to it, both in terms of your stock performance and your credit performance, any -- I think regrets is the wrong word -- but anything you would have done differently in terms of underwriting? Or are you still comfortable with the long-term viability of the opportunity?

It's hard not to feel chastened when your credit and your credit spreads and your stock price have moved this much in the wrong direction. Again, we are pretty much focused on the fundamentals, the deep dive, really understanding the assets. And from that standpoint, still quite comfortable. I know people are shocked by that, but we've spent a lot of time underwriting, and we underwrote to a downside. And as you know, we bought a company that we felt pretty comfortable that we could extract a fair price from.

The flip side is we also have an integrated team now that's raring to go. We've got eight offices. As the market has gotten more interesting, I think, we can cut them loose. We're still trying to work through exactly where the best places to deploy those resources are. But I think long-term you're going to see that be a positive.

Near-term, we want to go into every fourth quarter of every year with tons of liquidity. It's been our historical pattern to believe that that's when the market opportunities get the most interesting. And the one thing that the Fremont transaction did is it locked us out of the market for a large period of time this year, where we couldn't load up on liquidity because we had unreported material events going on in the Company. So probably my biggest regret is, unlike years past, we just didn't have a chance to reload; we didn't stock in a lot of liquidity in our unsecured credit line. We would have gone even deeper and taken down even more capital. You've heard us talk over the last couple of years of an impending issue in this marketplace, and we wanted to be perfectly positioned for it. And candidly, we're not perfectly positioned, but we're still pretty well-positioned. So, if I have a regret it's that we don't have a couple extra billion of capital lying around where we could really go to work on this market.

Jay, I'm wondering, given your core competency in what the Company does best, underwriting commercial real estate risk, have you looked at all at the CMBS market, the dislocations there, whether it's the derivative or the cash market? Maybe there are some opportunities for some very yieldy securities that you could purchase.

(inaudible) tell you, I still like underwriting assets. There are a few organizations out there who can underwrite 100, 200 assets at a time in a week and a half. We can't do it. We take six, eight weeks to do one asset. We put 40 or 50 different parts of our -- different people in different parts of our firm on every asset and every portfolio. We just don't have the scale to underwrite with the same level of confidence a security where you're making actuarial assumptions, statistical assumptions. We like to touch it and feel it and poke it and turn it upside down and walk the market. You just don't have the time or ability to do that. CMBS gets really attractive when it just gaps out. And right now, we'd not disagree with you, there are probably some really interesting opportunities. But it's not our core competency; it just isn't. And there's plenty of opportunities in our core markets. I don't think we have to stretch here to go try to do something that -- I think there are other people actually better positioned than us to take advantage of that.

Just a final question for Katy. Katy, you mentioned the unencumbered asset pool and the ability to tap, perhaps in lieu of going to the unsecured credit markets, or the equity markets, given where your securities are pricing right now. I just want to make sure that I am clear that that won't be viewed by the rating agencies as a reversal of course. Given that you've focused for the last several years on becoming an unsecured borrower, to the extent that you tap that unencumbered asset pool and borrow secured, even on an interim basis perhaps, is there any fallout from that that we should expect?

I think the agencies do look at secured debt levels. I don't think there is any particular percentage that they're focused on. But I do think that we have taken and enjoyed very much being an unsecured borrower with a lot of unencumbered assets. I think we're unusual today as a finance company to have the high level of unencumbered assets. Very few operate with single-digit numbers in that zone. So, I do think that the agencies take more than just one number or one unencumbered asset number into account in the rating. And I think they have a pretty good understanding of the flexibility that that affords us in the market. So I don't think they would view it necessarily as a reversal of course, but maybe more of a normalization, and actually demonstrating the liquidity of some of the assets in our book.

Just to follow-on, our AutoStar business can't fund itself. Our European business has access to capital. Our timber business has access to capital. Our corporate liquid securities business has access to capital. We have historically just simply done it out of one big pot. But, we will look at cost of funds across the board. As Katie said, this is still going to be a small, small fraction of the book. But, I think it's an appropriate thing. We have run in the past with small pots of secured debt. And I don't think it's going to be any surprise to them if we tap a few small pieces of the puzzle.

James Fotheringham, Goldman Sachs.

Jay, relative to your stated wish for more capital in the current environment, are you in any way tempted to reconsider the current REIT structure?

No. Never have been. I think long-term, this is still a very powerful business model. We are not feeling like we have to do anything materially different. The market has opened up to us the way we had hoped. There are going to be some really good investment opportunities. As I mentioned, Jay Nydick has the opportunity to raise third-party capital. There are plenty of people who want to invest with us. And if their opportunities just are so expansive, and our credit spreads and our stock price don't allow us to directly attack them, there are plenty of indirect ways to get the benefit of our skills and our platform and our knowledge base off of some other form of capital. And we will do that long before we think about something as drastic as changing our tax and corporate structure.

Understood. Just one other thing, just getting back to Dave's question on the guidance. I think I understand the very conservative assumptions that get you to your EPS guidance range for '08. But could you give us some idea of what this internal model, as I think you called it, is telling you? How far away is your optimistic scenario relative to this conservative guidance range that we have now?

I think on the asset side, we're pretty positive. And I think relative to where our asset spreads have been, and perhaps more importantly where structures have been, we think the market is definitely going to move there now. The corporate market repriced in the space of a couple weeks. It's taking the real estate market months just to take its first steps in that direction. So it's just a slower-moving marketplace. It looks a lot more like a sine wave. And you've got to get positioned. You've got to lay some quotes out there. You've got to be in dialogue with borrowers in multiple markets. And then it starts to happen, as the overall market starts to reassess where levels are. That's just now happening. So, again, I think, it would be imprudent for us to say we're going to wake up on January 1st and the whole market will have been repriced. We just don't believe that. But we are seeing signs that repricing is beginning.

I think the more interesting question for us, and probably the bigger switch for us, is this cost of funds issue. Historically, our speed, our ability to move quickly in the unsecured markets, has given us a material competitive advantage. If those cost of funds become available to us, as they have historically, even at slightly wider numbers, we are going to do a lot of business early in the year. But if you look at the current credit markets, and particularly our current credit spreads, again, we're scratching our heads. We're not sure why the concerns are being reflected in those spreads, or what concerns are being reflected. But the sooner those go back to what we would consider fair and normal, the faster you will see us committing capital, and the quicker the impact on next year's earnings. But, given where they are today, again, it wouldn't be wise for us to sit here and say they're going to snap back by January 1st, and here we go.

David Boardman, Wachovia.

You noted the dislocation specifically in the equity markets and in the unsecured debt market. You talked also about the shares you bought back here in Q3; looked like an average share price of around $33. I was wondering if you could talk about your expectations going forward on the share buyback, and anything you've already done here in Q4.

We've talked with our board recently at our quarterly board meeting. There's no question we all believe the stock is compelling; the preferred is compelling; the bonds are compelling. There are plenty of pieces of our capital structure that we know in our heart of hearts are wildly mispriced right now. The question is where do we most effectively spend our time, our effort and our capital? We have 2 million shares left on the buyback. That's not going to change anything in our company. It may turn out to be a very attractive investment, and we'll evaluate it on that basis. If it's the best place to put money, we're going to keep putting money there. But I would also tell you that long-term for the franchise value of the Company, I'd rather make 15% on a great investment or in a new interesting area than just buy back stock. So we're going to be very judicious and prudent about it. Clearly, it's an attractive investment today, and clearly, you'll likely see us continue to buy if the stock stays where it is since we were buying at 33. But whether we're going to go beyond that, and how much we would go beyond that, we're going to stay in dialogue with our board and watch where the market opportunities present themselves.

I was wondering also if you could compare and contrast what's been going on and what's going on right now in the US versus Europe, and how you're positioned maybe to take advantage of what might be going on over there as well.

I'll be honest; the best opportunities right now are in the US. The most dislocations are in the US. Europe is grinding. I think they're still trying to find a level. The opportunities are getting a little more interesting over there. But as we have historically said, our focus over there is to team up with really great partners and find those anomalies in the marketplace that we can take advantage of. I wouldn't say the overall correction over there makes me more compelled to move abroad; in fact, I'd say the relative opportunity set in the US is so good, we're going to be staying here with most of our capital for a while.

Just to revisit quickly the discount, and how that's going to be amortized into income. I think, Katy, you had said about 100 to 120 million of expected loan loss reserves going forward. Is that about the amount we should expect of that discount that will be accreted into income going forward?

That's what our model and sort of the earnings guidance that we laid out for you was predicated on. But, obviously, we set our reserves quarterly based on the quarterly risk rating assessment. So we don't expect there to be a straight-lining, or an actual number every quarter that we use. We really look at each asset that's on the NPL and watch list. We look not only, as Jay mentioned, at all the numbers and all the actual data that we have, but then we use some judgment in determining how we think those various outcomes will come into play. So it's a little more complicated than just sort of straight-lining reserves. But we do think they'll probably be a little more heavily weighted in the next couple of quarters as we continue to look through the Fremont portfolio. As you know, Fremont -- the weighted average maturity is fairly short. So we do anticipate having a little heavier reserve volume in the next couple of quarters related to Fremont.

Susan Berliner, Bear Stearns.

I was wondering if you guys could just go over -- it seems like you changed around your collateral type, obviously, with the inclusion of Fremont. I was wondering if you could kind of go through there and give us overall what your residential condo plus land would be. And then secondly, on your assets held for sale, the 77 million, I was wondering what those were, and if there were other asset sales you could look to do instead of trying to raise money.

I guess there's two big changes on the composition chart. One is the breakout of corporate transactions. As I mentioned, we saw some pretty interesting stuff in the third quarter. Working together with Oakhill on a lot of opportunities, we saw some of the best risk-adjusted returns there. That number has gotten big enough. We decided to pull it out of other and drop it into the chart explicitly, breaking it into corporate that has a real estate component that relies a lot on kind of iStar underwriting. And then, I think, we had about 2% of straight corporate that were just compelling risk-adjusted returns that Oakhill kind of said, look, this is the best market we've seen in a long time, you should buy a piece of it.

The other big change is the breakout of land, pulling that out of apartment residential. A lot of the transactions that Katy mentioned are just never going to get to the finish line or didn't meet their presales are no longer going to be condominiums or multifamily, they're just going to be land. And adding that to the iStar land also made it large enough we needed to break it out. So with Fremont's land component, our land component, and now a lot of assets that we thought were going to be apartment residential dropping to that category, what you saw, which was interesting to me -- I didn't expect it -- was apartment residential actually going down, even though the preponderance of Fremont's portfolio was apartment residential in our initial categorization. As some of those have dropped out of that category, apartment residential actually shrank, and land went up enough that we thought it needed to be broken out.

With respect to the asset sales, we actually had a few very small asset sales this quarter. I think they totaled like $5 million. We had a gain on those of about $1 million. And then, as you know, we do continue to look at our CTL portfolio and prune assets when we believe they are nonstrategic or it's appropriate. And we do -- we are in the process of selling an asset that's located out in the Silicon Valley that we've held for a while, and believe that the valuations now are coming back to a point that makes sense to probably sell it.

I guess the other thing I would say is the most liquid part of our portfolio is this corporate liquid security portfolio. That's kind of instant on liquidity. So, if we saw an opportunity that was compelling, that's the place you can go to first and fastest.

David Fick, Stifel Nicolaus.

Can you talk a little bit more about the progress of the Fremont integration -- obviously, you have to be cautious on a call like this, but -- where you stand with the people, and sort of recycling their psychology into the iStar way?

As I mentioned, the thing I am most happy about is the immediate access to their market knowledge and their relationship base and customer base and brokerage relationships. And that's already coming to bear. I think they, like us, are a little bit frustrated here. We want to cut loose on this marketplace. But we're pushing them to get not only paid for credit risk, but also paid for liquidity. And that's something that even inside our own organization -- it's a little bit of a shock to a borrower who for the last five years has basically gotten liquidity for free, free extensions, free everything, to now being asked to look at it and say for the provision of liquidity we want to be paid, and for the underwriting credit risk we want to be paid. And that transition is taking place not only inside, I think, their thought process, but also in our own. So, that process is working its way through. As they get a couple of deals all the way through from start to finish, I think, we're continuing to all come to the same page. And I think it's going to be a very powerful organization in terms of our local market presence and our kind of national relationships out of New York. We still feel really good about that.

On the other side of the equation, kind of the asset management, cash management, some of the back-office stuff that we never talk about on these calls, again, we've materially upgraded a lot of the pieces of the puzzle, the systems that support the kind of growth we've had and the kind of growth we expect, the breadth and expansion both internationally and across new platforms. All of that is getting -- little pieces are getting filled in. We gave ourselves six months to kind of do nothing on the investment side; just focus on the integration. We've got a couple months left, and I feel pretty confident by the beginning of January we're going to have a team that's ready to go on almost all fronts. And that was our goal, and I think we're still on track to achieve that.

As you continue to come through the Fremont portfolio, what are you finding that surprises you?

Got to get rid of the little loans. I think there's almost 100 loans that are under $10 million. And when you sit through a 650 asset review, and 100 of them are 3 and 4 and $5 million deals -- historically, we talk about markets and sponsors and -- it's just not worth it. So I think we were surprised how much extra time and effort we have had to deploy to get our hands around even 3 and $5 million deals. I think we're past that now. I think we feel a lot better about that.

I think the other thing that everybody in this room feels and has felt is when you go through the whole 650 assets in the entire company, you leave the room feeling really good. But when you sit in an asset management meeting and you go through 35 assets, every one of which is bad, you feel like, oh my God, what did we do? And I think you have to keep some perspective here. 90% of this book is really good stuff, most of it either at par or worth well more than par. It's that 10% we're spending -- the old management rule, 80-20 -- we're spending 80% of our time on 10% of the portfolio, and trying to keep everybody focused on the big picture, which is we've built a great portfolio; we have great diversity; we've got tremendous capital strength and flexibility. And we bought this portfolio at a discount. We should expect to take $280 million of losses, and we'll still make a 20 ROE. Sometimes you lose sight of that forest for the trees when you're slogging through the borrowers not paying, thinking I got to work this out. But a couple of us have been through some cycles; we've worked out a lot of deals. Personally, I find it one of the more interesting ways to learn about how real estate really reacts. In the last five and seven years, people haven't really had to think about that. I'm happy our firm gets to see up close and personal exactly what happens in a workout, and where those legal documents really can protect you, and how borrowers really do act when they have a guarantee on the line. Those are things that it's important for our junior people to see, and, frankly, for everybody in our firm to see. And as long as it stays within that $280 million loss basket, this will have been a very good thing for this firm. So our job is to make sure we don't get surprised and nothing gets outside that basket.

Back to the big picture on Fremont for a minute, one thing that has shifted dramatically since you committed to this transaction is, obviously, the overall commercial lending world. With respect to takeout assumptions, virtually all of these loans are short-term; virtually all of these loans have a period of less than 60 months, let's say, going forward. And so, we are going to have to find an alternate source of financing. What are you now assuming about even very good assets that may have difficulty financing out?

Great question. It's one we've spent a lot of time on. Let's go back to that example I mentioned. There's a $100 million loan. It's going to be built over 24 to 30 months. That capital structure probably has a mezzanine person in there and some equity in there. In a lot of cases, those deals are at 65, 70% loan to cost. Those are the ones I'm really not worried about, David. Because the borrower spent $100 million thinking they were going to make 130 million. And they've got 20 million to protect, and the mezz guy's got 15 or 20 million to protect. You've got to go from 130 million down to 50 million for us to fall outside that $280 million loss bucket. I am very comfortable, as I hope you are, that if you built a good quality project in a good location, I'm willing to wait. And I think the mezz guy's probably willing to wait. And in many cases, I think, the equity guy is willing to keep putting up money to wait. But they're just not willing to sell a brand-new steel, glass and concrete tower at $0.35, $0.40 on the dollar. And I fundamentally -- and maybe I'm wrong -- but I don't believe that that's a scenario we're going to see again and again across the country.

I think the ones you spend time on and the ones you worry about is it's coming online very soon, there's not a lot of sponsor capital or mezzanine capital to protect it, and the sponsors are going to look you in the face and say, look, I've got all these presales, I don't think they're going to close -- what should we do? Our first order of business is to get them to take the risk. They got the upside; they should take the downside. But when they don't, we are telling them hand me the keys. So a lot of the NPL list is stuff that we think either we'll get our money back, or we'll take a loss within the range we think. But we're not blinking just throw it on NPL and tell the borrower if you don't pay us it's game over.

Now, when they treat us well, we're willing to work with them. And we have counseled a number of them -- look -- you've got somebody who has got 20% down on a deposit, give them another 10% off, give them 15% off. It probably means you're losing equity as you go, but we're telling you that's a better scenario than waiting this out and ending up with nothing.

So some of the deals -- stuff is still closing, even in south Florida. There are assets still closing deals where people have 20% down. And I think for the next six to 12 months, we're still comfortable that a lot of these positions are going to work their way through. It's really the end of '08 and early '09 when the stuff that was modeled at the beginning of '06 comes online at probably the peak of the market pricing. You better be scrambling right now. You better be working with those customers. You better be offering them some inducement to close. And honestly, if those come back to us in '08 or '09, nothing is getting built for '10, '11 and '12. And if we have to carry an asset for a year, we're a $20 billion company; we'll carry an asset for a year. It's just not enough of them that I wake up in the middle of the night and go, oh my gosh, we have hundreds of assets that have problems. We just don't have that.

There are going to be some. We are going to lose somewhere between zero and 280 million. That's our prediction. And we underwrote it to make a 20 if we lose 280 million, and we are still comfortable with that number. Might be wrong; could get worse; but it's not going to get worse tomorrow. It's probably going to be '08, '09 before you go, okay, we were a little bit wrong. And by then we will have, I hope, made so much money on everything else that came out of this transaction that we will be able to cover that, too.

What is your underlying assumption about demand for commercial and residential real estate in that statement? Do you assume a US recession?

No.

What if there is a US recession?

Tack another year on. Here's the good thing. I want there to be no new home sales, right now. I want the world to suffer all its pain right now. So, the faster and harder this market corrects and gets rid of all the excesses, the better for us. Because most of our stuff is not coming on right now. The commercial side of our book is really solid. So I don't lose any sleep over that. It's the residential side. And I think that's a matter of time. I think most of the projects continue to be in good locations and have long-term viability. But as you said, if it comes on tomorrow, it's probably sitting for two years. And somebody's got to carry that. Now, I think the equity guys in most cases are going to take the first year. The mezz guy's going to take the second year. And it's really going to be our problem in '09. I'm happy to work with borrowers and give them every possible opportunity to prove that they believe and they want to stay in the game. But when push comes to shove, we have, as Katie said, dozens of construction managers, asset managers, property managers -- this is what they've done in their lives. We're not going to be afraid to take assets back. That does not mean we're going to lose a lot of money. And that's, I guess, the thing that maybe the disconnect is. We're sitting here looking at what happens to those assets that come online in '08, '09? We don't know. But I've got to tell you, I know I'm way below replacement cost in a market that supply has stopped two and a half years ago, where demand, just demographically, is going to be there in '10, '11, '12. And we're just trying to figure out how much should we mark on that particular asset? Trust me; we are taking marks; we are building reserves; we're going to be ready to say, okay, we're not going to get all our principal back, we're going to get most of it back. You've lived through enough of these cycles. Even in the worst cases, to lose $0.50 on a first mortgage on a brand-new building in a good location, that's really hard.

That's the RTC all over again. My last question relates to Katie's commentary on the non-funding significant portion of the original commitment. Where did you sort of reach the conclusion that you were going to have a significantly lower actual committed forward Fremont unfunded forward funding? And what costs are there associated? I assume you have some investment and some fee negotiation and so forth that goes on with abandoning those deals.

(Yes, it's asset-specific, and this really was a result of our drill-down on all of the assets -- where we have early-stage projects that have very little -- for example, the equity has funded, and maybe the mezz has funded, but our first mortgage has not begun funding yet -- we are in contact with those borrowers. And to the extent the project does not make sense in today's market, and they decide not to pursue the development currently, or to the extent we have, which we do in almost all of our documents, presale requirements or milestones for condos, if those milestones haven't been met, we believe, and in discussions with borrowers, that we will not be funding those particular mortgages. So those projects won't move forward.

What happens to the commitment fees? And what costs do you have associated with that?

It depends on where they are. It's a deal-by-deal discussion. It depends on where the borrower is.

They've already paid the fees.

In most cases the upfront fees have been paid.

We don't give those back. I'll give you one example. In our own book, an NPL that just came off the list in the last week was a $60 million construction loan with preleasing requirements. Somebody else came in and took the loan [audit]; they had missed the milestones; we weren't going to fund it. That's $40 million out of a committed basket from iStar that just disappeared. It's never going to be funded.

If these deals are well structured -- and to their credit, I think, in many deals Fremont actually used very tangible milestones, and said if you don't do this -- and I think they have a great construction group that understands what it takes to build assets -- if you don't do this we will not fund. Now, you don't use that as a technical reason to walk away from a good customer, but you hold their feet to the fire and say if you change the risk on us, we're going to have a conversation. In some cases, we're not funding at all because your business plan has fallen apart. In other cases it just changes the risk structure and it's time for a new dialogue. But as I said, we try to work with our customers, be reasoned and thoughtful. But if you've changed and the world has changed, we're going to have a new conversation.

Mike Swotes, Angelo, Gordon.

I think it's been about 5.5 years since your stock last traded at these levels for any period of time. I was wondering if you could talk about some of the major differences between the iStar 5.5 years ago and the iStar today, and maybe comment on your competitive positioning and ability to withstand some of the turmoil that we're seeing.

I hate to revisit those days, but really, after we bought TriNet in the public market in 1999, our stock went through a similar decline. Our credit spreads widened out similarly, and to the same kind of position we find ourselves in today; every part of our capital structure looks compelling. And so I think the dynamics back then, at least from a technical perspective, weighing on the Company, are very similar. But as you point out, this is a fundamentally bigger, stronger, deeper, broader company with very strong capital base, great depth and penetration on key markets that right now are offering very attractive risk-adjusted returns, and an organization and a management team that is night and day from then.

So, yes; we're bigger and stronger. But we are still susceptible to market forces. And it's unfortunate; it really does feel a lot like that same period of time that no matter what we did, you just have to keep grinding out good returns, and the market will come and begin to respect that again. And I fully believe that. Our goal right now is to grind out very strong returns to show the depth and breadth and strength of the Company. On the capital market side we have tremendous flexibility, as Katy said. We never had that before. To me that's the real underpinning of why this company is stronger and better today, in many respects, than we ever envisioned.

When did you guys become investment-grade?

2004.

That wraps up our call today. I just want to thank Jay and Katie and everybody here, as well as everybody joining us this morning. If you should have any additional questions on today's earnings release, please feel free to contact me directly here in New York. [Kent], would you please provide the conference call replay instructions once again? Thank you.

I certainly will. Ladies and gentlemen, this conference will be available for replay starting today, Tuesday November 6, 2007 at 1:30 PM Eastern Time, and it will be available through Tuesday, November 20 at midnight Eastern Time. You may access the AT&T executive playback service by dialing 1-800-475-6701 from within the United States or Canada, or from outside the United States or Canada, please dial 320-365-3844, and then enter the access code of 891239. That does conclude our conference for today.