ACAS' Exits: An Update

In ACAS' recent announcement of its Narus exit (through transaction yielding $21m to ACAS and its controlled funds, including $12m to ACAS' own top line and $3m to ACAS' bottom line), ACAS included a link to a non-exhaustive page of exits. Presumably, this non-exhaustive page of exits reflects the deals that are material and aren't subject to nondisclosure agreements (some buyers may pay for silence), but if you sort the list by date of exit you'll notice that it hasn't been updated since October of 2009.

Earth to ACAS: as of this writing, that list is close to a year out of date. Inquiring minds want to know, and it's an obvious place for folks checking up on the company's results to look. To avoid looking like ACAS is ashamed of recent results, ACAS should publish them.

Reviewing the list, it's clear that ACAS is willing to confess bad bets publicly (the investments in Stein World, Flexi-Mat, S-Tran Holdings, Weber-Nickel Technologies, and Sunfuel Midstream range from 90% losses to genuine 100% loss), but that some of ACAS' biggest listed investments have had attractive internal rates of return (Extream Software at $548m, Axygen Bioscience at $271m, Evans Analytical at $125m, and HomeAway at $120m had IRRs of 22%, 22%, 80%, and 39%, respectively). Some of the investments that involve well-known names haven't been bad: Piper Aircraft's $91m investment produced an IRR of 19%, and Gibson Guitar's $33m yielded an IRR of 16%. But where's Riddell? Maybe Riddell isn't listed because ACAS kept a couple percent, and is therefore not entirely exited.

The vast majority of ACAS' deals aren't on the web site cheat-sheet. Not appearing anywhere investors can see, though, is something even more important: the vast bulk of deals on which ACAS takes a pass (about which one can read here). To make all that proposed-deal volume valuable, ACAS needs effective screening. The fact that there's a huge volume of prospects is valuable only to the extent ACAS can tell gems from duds. The firm's history of buying investments under conditions designed to result in profitable exits even as multiples contract shows a longstanding plan for conservative purchase, but the recent downturn has rattled confidence in management's ability to pick winners.

Continued exits from investments – especially loser investments – places ACAS in a better cash position (a) when a few hundred million cash will reduce ACAS' interest rate on its entire debt, and (b) just at the time that private companies unable to access capital from gun-shy banks and other sources of funds are even more on-sale than normal due to the P/E multiple contraction, and presumably therefore offer some of the most promising prospective investments. And then there are distress opportunities, in which ACAS has been apparently investing already.

While the risk of double-dip recession remains frightening, the ability to get good deals in purchases of solid businesses seems very attractive: with so few equity buyers, ACAS may be the only game in town, so to speak. The more deals ACAS can review, the more likely ACAS should be to find outstanding winners.

I'm off to do some of my own work, and I'll leave management at ACAS to keep my capital there hard at work while I'm otherwise occupied. The dividend won't exist for another couple of years, I expect, and this cash retention will help ACAS grow NAV and, as the capital base grows, ACAS' profits. I like the long term now as much as ever.

The next couple of quarterly reports should be interesting as we see ACAS' activity to continue improving debt cost and directing its attention away from its own finances and toward the books it needs to review to make good deals. And that, folks, is where I want ACAS' attention directed.

Consumer Update: Bank of America

After having spent over a week getting Bank of America to begin automatic, recurring payment of a home mortgage recently bought by Bank of America, I have an update:
Nice, eh?

I discovered this while on the phone with a customer service representative who called me in response to by scathing feedback in one of several surveys I was invited to take in response to my multiple, fruitless contacts with Bank of America to set up automatic recurring payment of the mortgage. After 14 minutes on hold, she tells me that (a) according to the mortgage department it is set up to pay the mortgage automatically every month and (b) if I will write down a host of access numbers, pin numbers, etc. I can make my next call quicker and the options offered by the automated phone system more relevant. What she says in (a) may be true, but the BS meter goes offscale on (b).

Much more of this and I'll refinance with another bank just to be rid of the bastards.

Bank of America: Don't Bother

After deciding not to borrow from Bank of America, I ended up with Bank of America's "service" anyway: they bought my loan.

So, to avoid the SNAFUs I foresaw with Bank of America screwing up ordinary transactions as I'd seen in the past, I decided to make timely payment of the loan Bank of America's responsibility: I decided to set up automatic recurring payment of the loan. The steps to reproduce my experience are:

(1) Go through the dance to set up an online ID for managing Bank of America accounts. There were several steps, involving emails and snail-mail instructions, but this was ultimately accomplished without incident. Ameriprise Bank could learn something.

(2) Search diligently for evidence of how Bank of American invites users to accomplish payment online. This can take a while. The online information does indicate that setting up what Bank of America calls a "pay plan" – the automatic recurring periodic payment of an amount due under a loan payable to Bank of America – can be accomplished online. Searching for the mechanism to accomplish this will take longer than your patience endures, however, because establishing automatic payments of amounts due under mortgage agreements has not been possible since last October. This last but critical fact is not disclosed online, and is entirely inconsistent with – in fact, directly contrary to – the instructions advising users they can establish pay plans online.

(3) Call for help.
(a) This part is tricky because the numbers to which mortgage customers are routed tries very hard to steer you into making a one-time payment, and there is no recurring payment mechanism available by phone. This mirrors the situation online, but it's easier to identify on the phone because the options are fewer in number. When you try pressing "0" to get a human, a voice warns you that using a human can lead to higher fees. It is possible, using contact numbers buried in the web site, to get a human without facing the phone queue and its fee threats, but this is probably not the experience you'll have from the outset.
(b) Let's say that after an hour of trying to follow web instructions that don't work, you eventually get a phone number that leads you to a human. This is where, in hindsight, things really because irritating. The human was obviously consulting notes in giving his instructions, which lent an air of authority: he wasn't making this up, he was guided by printed materials to keep everybody giving accurate information. He instructed me that I needed to set this up online. You are going to see the punch line coming long before I did on this one, but bear with me. I was carefully led through a laborious process of associating my external "pay from" account with my online mortgage account so I'd be able to set all this up, and even informed that I was just one step away from completing the automatic payment setup.
(c) I entered bank info online so BoA could pull cash from my account at will, and I looked for a way to ask it to start. The helpful guy told me that, of the options I read him, I needed "verify" – that this would clear the account to start automatically paying the mortgage bill.
(d) Over the next days, I read from my account at a real financial institution that Bank of American had deposited $0.43 and $0.63 into my external "pay from" account. I dutifully logged into the BoA site and entered these numbers in the account verification boxes. My external "pay from" account was now verified! Hurray! I was done! And only three days of effort!
(e) Or was I? I tried to find evidence the account was actually going to be charged. The confirmation information simply said the account had been verified and that it was ready for making payments, not that any were scheduled. The online system included a make-a-payment link for folks who wanted to pay from an external account (or another BoA account), so it was quite plausible that nothing was set up at all. I began hunting for evidence of how to start the automatic payments. See Step (2) above. After patience failed, I began Step (3) again, looking for a human to finish what was obviously half-done.

(4) Get Help For Real. The same phone queues I accessed in (3) above tried to direct me to make one-time payments and actually hung up on me. In my last call, after 37 minutes of alternately being interrupted by agents trying to tell me that "bill pay" can't be used on a mortgage (BoA has a product called "bill pay" that is used for something else) and being placed on hold waiting for someone able to speak intelligibly about the bank's products, and being placed on hold waiting for someone authorized by BoA to confirm I intended to authorize BoA's payment of the mortgage from the account I verified online and from which BoA presumably would allow me to make an unlimited number of payments in any amount I chose, I finally was told the mortgage bill was set to be paid on time every month without one of the fees BoA charges for many of its payment options (the payment scheme is dizzying, and every tier has a different fee).

All told, it was about a week to get the account set up to pay the mortgage loan automatically, a half-dozen calls, numerous trips to the web site, and lots of mistaken online and human-provided instruction. An enormous waste of time.

I'll be looking for alternatives to end up in business with a different bank. As for you: borrow from someone else. Try to make sure you borrow from someone who'll be servicing the loan, not just originating it. You may save some middle-man fees that way, too.

On ACAS following 2Q2010

Thanks for the Anonymous post trying to point out the weak parts of the ACAS story – the more people have to think about, the better. ACAS' 2Q2010 presentation is worth looking at – the slides, at least, if you don't think you have time for the whole presentation.

I do not suggest ACAS should refinance immediately – refinancing was, after all, an expensive undertaking and I expect ACAS to enjoy being out of the refinancing market for some time. However, over the long run, it's an issue that will recur due to the restriction in the new debt limiting the use of funds raised by subsequent debt. I don't see this being an issue before the 2012 time frame. I see this being an issue when ACAS is wanting to be more levered than it is. At present, achieving ACAS' maximum allowable leverage doesn't involve de-levering.

ACAS won't worry about that soon. Looking at 2Q2010 presentation slide 15, ACAS also plans achieving leverage in the future using on-balance-sheet securitizations, a strategy with which ACAS' management has apparently had both experience and success at its publicly-traded controlled fund AGNC. Thus, ACAS may seek future leverage with investment partners on terms that are more attractive than generally available from banks.

Cash flow isn't operating income. ACAS' realized losses washed out its NOI while providing positive cash, preventing a dividend-related liquidity crisis in the near term, but the accounting profits (taxable income) at portfolio companies that lay behind the NOI figure should not be mistaken for positive cash into ACAS. ACAS is paying lots of interest on the debt whose nonpayment is still mounting. To combat this, ACAS plans improving its margins by dropping its interest rate another percentage point by the end of the year through repayment of secured debt. (see 2Q2010 presentation slide #15) (Note that Anonymous seems to overstate the restriction on repayment; the make-whole payments protect only a small subset of the new debt, most of which is subject to repayment at will without penalty, and some of which has a mandatory repayment timetable within the time frame described by Anonymous. If ACAS makes its planned early repayment, it will do so without penalty.) NOI is just coming off cyclical lows, and should improve – and with it, the internal metrics of the companies providing the NOI.

On the other hand, although nonpayment is worse sequentially – nonpaying loans have grown from $671m cost at the end of 1Q2010 to $686m cost at the end of 2Q2010 – non-accrual has improved y/y from $996m cost on non-accrual in 2Q2009. "Past due" loans stand at $57m (cost), down from a peak of $209m at 3Q2010. The benefit to ACAS of having a non-accruing loan that isn't past due seems somewhat mysterious to me just now, though. The fact that ACAS makes a deal for forebearance to allow a portfolio company to reinvest instead of pay interest doesn't mean ACAS ends up with either operating income or interest payments; it means ACAS is helping portfolio companies to avoid the vortex they're circling. It might promise some better odds of future recovery, but from where we stand out here it's hard to measure that benefit. Still, ACAS' focus on organic growth within portfolio companies and in supporting add-on acquisitions suggests that ACAS' efforts to support portfolio companies in their time of need has a strategy beyond improvement in the prayer life of executives. (See 2Q2010 presentation slides 20 & 23).

I note that although 2Q2010 has significantly more NOI than 2Q2009, its NOI was sequentially down from the 1Q2010 quarter. The NOI decrease could be partially explained by the increase in non-accruing loans over the quarter, but this isn't all of it.

The upside is in two hard-to-predict parts: recovering investment values and evaporating NAV discount. This is necessarily a lumpy ride. Depending on the economy (operating results) and the markets (multiples), we could have some significant backward movement while we're underway. Comparing the realized earnings to the FAS-157-compliant SEC-reported "earnings" shows ACAS negative except in improvement in unrealized gains.

The BLTs don't seem to be performing as planned at all. Repayment of debt associated with the superior debt classes issued by the BLTs is something I haven't measured, but the "half ACAS' cash" conclusion may indeed be the number. The opacity of ACAS' CDO investments prevents me from making any intelligent statements about them, but I notice that management has long since stopped using the CDOs as an example of how FAS 157 misprices ACAS' assets. Conclusion: management agrees that the CDOs (like the commercial mortgage-backed security investment once touted as expected to perform through maturity, despite being valued poorly as residential CDOs were collapsing) are essentially worthless.

To the extent ACAS needs cash, ACAS will be depending on exits. Fortunately, ACAS has been strong on making exits. Moreover, ACAS' focus on debt and securitizeable mezzanine business going forward would, if successful, result in quicker exits and less equity exposure. (see 2Q2010 presentation slide 18) With respect to "earnings" though, ACAS' reversal of unrealized losses and its increases in unrealized gains will likely keep it (and its NAV) looking good in its quarterly reports. However, as pointed out by Anonymous, most of ACAS' writedowns have been in equity. Most of the exits, however, have been in debt: of $351M of 2Q2010 cash realizations, only $45M came from equity. The rest has been debt. Retaining equity that stands a fair prospect of turning around with the economic recovery is attractive.

As depicted on 2Q2010 presentation slide #9, the sale of assets has been near recently-reported fair values. 2Q2010's exits at 7.7% above NAV are better than the within-2%-of-fair-value amounts that have characterized exits since 3Q2008, and it'll take some time to ascertain whether this represents a fluke or a trend. In the immediately prior quarter, exits were 3.7% below last-reported fair values. Both numbers are likely pulled off parity by a small number of outlier exits, and may be useless to shareholders in estimating the value of likely future exits, other than to reassure them that exits are plausible near reported fair values. Since management wants more than reported fair values from some of its investments, management will likely hold on to some well-performing investments for some time.

On the other hand, some of that equity isn't coming back: it died. ACAS borrowed money and lost it in deals that went south, and will have to repay it with the profit from its surviving deals. We won't be seeing $40 shares of ACAS any time soon.

But, as suggested by Anonymous, seeing $15-$20 in a handful of years is plausible. In my view, it will turn largely on macro-economic conditions that impact the entire market's profits and multiples. Considering that ACAS' 22% growth in NAV came after the effect of dilutive issuance, and that ACAS continues to hold a significant fraction of equity (including well-below-NAV-valued ECAS), the prospect for significant equity-driven return seems good – provided ACAS manages its liquidity concerns.

Reducing average leverage to 0.6 to 1, dropping interest rates to their lowest tier available under the newly-refinanced agreements, and lowering the principal on which debt will be paid will all reduce forward risk and improve forward margins. Raising new managed funds will enable ACAS to realize fee income without holding risk, allow ACAS to leverage due diligence overhead and internal resources to develop income beyond that achievable with ACAS' own funds, and enable ACAS to contract to enjoy potential upside like upside participation when return goals are met. Securitizeable mezzanine finance will both shorten the investment cycle and de-risk the portfolio. Focus on growing existing investments organically will allow ACAS to invest in what it knows, picking well-understood known winners for add-on investment in anticipation of preparing them for resale as they come off cyclical lows both in internal metrics and in terms of external multiples. Although ACAS is currently free under its debt agreements and the BDC Act to declare and pay dividends, the lack of an apparent need to declare a dividend this year or next will allow ACAS to maintain and recycle capital on hand for the near-term future.

In the more distant future, ACAS may market itself as an income stock, but in the meantime it's a holding company with no tax expense and I look forward to observing its performance improving book value. I'm optimistic that NAV improvement will continue as the economy avoids further collapse, and as ACAS proves its capacity to do profitable business over time its NAV discount may dissipate. Even if it doesn't, recycling funds into high-value new opportunities created by the economic collapse will offer returns by which ACAS will reward shareholdes with (ultimately) statutorily-mandated ordinary income distributions in the form of annual dividends.

ACAS: 2Q2010

American Capital recently announced its results for the second quarter of 2010, and those results are encouraging. In reading them, consider that American Capital spent the quarter threatening a bankruptcy filing to encourage creditors to accept a debt refinancing deal, which meant that in addition to suffering ongoing distraction at the top of ACAS as it renegotiated its debt, ACAS also had to spend money hand-over-fist on bankruptcy specialists so that it could, on immediate notice in the event deadlines lapsed, file in a bankruptcy court to prevent creditors from asserting acceleration and other rights held under the old debt agreements, under which ACAS was to have maintained a net asset value that proved implausible as the financial markets imploded in 2008. The expense of these renegotiation tactics was $17m over the quarter (up from $7m in the year-ago quarter). The end of the debt renegotiation battle means (a) less distraction for top management, and (b) decreased expenses.

Expense Reduction = Income Increase
To grasp the significance of the decrease in expenses, one should consider ACAS' net operating income (NOI). This is the income gained from operations either because portfolio companies profit (which profits would flow to ACAS' bottom line in the case of the mostly-owned portfolio investments), or because ACAS receives payments on debt extended to portfolio companies (ACAS profits to the extent of the net of its debt income and its debt expense; much of ACAS' lended funds are themselves borrowed). NOI includes everything but investment performance (that is, the change in market value of what ACAS owns); it's the income ACAS makes standing still and doing no deals. And what was that NOI in Q2 of 2010? ACAS' NOI was $29 million over the quarter. Doing nothing else to ACAS' business and merely dropping the debt renegotiation expense, ACAS' NOI would increase over 58% to $46m. This gives one an illustration how significant the elimination of the debt refinancing problem is to ACAS' ongoing operations.

ACAS' NOI hasn't been standing still, though: the current NOI represents a 45% increase over the NOI it posted in Q2 of 2009. As the economy behaves in a more normal fashion, ACAS' portfolio companies will be able to operate in a more normal fashion, and the results of those normalized operations will flow through to ACAS in the form of resumed debt repayment and improved business operations within portfolio companies. The Jaded Consumer has long advocated looking to NOI for an indication of the success of management in picking winning investments, and it's good to see that number moving up -- and looking to continue moving up.

The downside on the NOI is that at $0.09/sh, it is exactly where it was a year ago – that is, share issuance served to dilute NOI performance per share back to its year-ago levels. But there's an upside: even as NOI has stayed the same, ACAS' performance other than in operating income has moved forward.

NAV As Barometer of Management Performance
Even as ACAS posted $0.09 in NOI, it has grown net asset value more than the $0.09 one would expect in the absence of a dividend. NOI increased $0.17/share to $9.15, a 2% increase from the prior quarter. Compared to the NAV announced at the close of 2Q2009, ACAS is up over 23%. That might sound good, but the truth is better. Most companies grow their assets by posting profits on which they accrue tax liabilities. ACAS has achieved its recent results while making investment exits at a net loss – protecting it from being forced to disgorge assets either to the government as taxes or to shareholders as dividends. (As a BDC, ACAS' dividend payment is statutorily set within a range that is a function of ACAS' taxable profit; since FAS 157 and the SEC-reported "income" isn't taxable profit, those aren't the numbers one looks to either for tax liability or for dividend payment. ACAS will have no need to lose valuable cash to dividends for the near future, and will have no tax liability either directly or as passed to investors.) Since much of ACAS' portfolio is valued on the basis of things like price-to-earnings multiples (that are constrained under current markets, and would be expected to expand over the long term as macroeconomic conditions make investors more excited about equities as compared to fixed income investments) and earnings themselves (that are adversely impacted by things like unemployment and spending reductions that accompany the current economic cycle, but are cyclical and absent business failure would be expected to recover), the re-inflation of ACAS' NAV is likely only just begun. As ACAS begins turning in taxable profits, ACAS' loss carryforward will stave off tax liability for maybe another year or so.

When Earnings Aren't Earnings: What to Look For in ACAS
In the year-ago quarter, ACAS' FAS-157-compliant "earnings" were ($2.52) per diluted share (2Q2009), but the newly-announced quarter shows a gain of $0.84/share. Again, this emphasizes the disconnect between FAS 157 and the real world of taxable gains: ACAS is allowed (nay, required) to tell the IRS it lost money (and owes no tax, and has a loss carryforward that is growing so as to prevent near-term dividend issuance), but at the same time must report to the SEC (and to you and me) in accordance with FAS 157 and proclaim it's made 84¢/sh for the quarter, which at recent prices of $5.30 or so a share, suggests a miniscule multiple of price to earnings.

The tax loss means that dividend-paying requirements are pushed further into the future, which is good because ACAS pays good money for all its cash, and should be making a return on that money whenever possible instead of mailing it to speculators. I'd rather see the share price upward, thank you.

Overview
What does the 2Q2010 announcement tell us?

(1) NOI is set to rocket. NOI will swell with the decreased debt expense, which is not reflected in 2Q2010 even though the refinancing closed in the quarter. This is because the deal closed so late in the quarter that NOI reflected chiefly the default-rate interest ACAS was paying prior to the refinance. Because NOI includes ACAS' spread between its debt expense and its interest receipts, dropping its debt expense by reducing the principal by over $1B and by lowering the interest rate from the double digits to the single digits has a multiplicative effect on ACAS' NOI. NOI will also just stop being dampened by things like debt refinancing expense. Add in the elimination of the $17m debt refinancing expense, and it's clear NOI is sitting on a rocket for the next quarter.
Well, barring financial catastrophe that would affect portfolio companies' capacity to repay debt.

(2) ACAS won't be paying a dividend soon. This good for investors who want ACAS to get the most return out of its interest payments; why pay interest on money it has to distribute to shareholders and stop getting a return on?

(3) With rocketing NOI and no requirement for a dividend, ACAS will be generating lots of investable cash that (a) will lower ACAS' debt/equity ratio, (b) will have no interest expense, and (c) will multiply ACAS' returns as management seeks out more opportunities that are as mispriced as ACAS' own shares.[1]

And just how mispriced are ACAS' own shares? ACAS now trades at something like $5.30. At the end of the quarter, NAV stood at $9.15 and rising. This implies a NAV discount exceeding 40%. That means that if ACAS stopped working and focused its efforts on conducting an orderly liquidation, ACAS' shareholders would end up with a gain over current prices of over 50% (shares bought at $5.30 would result in a final dividend on dissolution of over $9, nearly $4 gain on a current investment). Since ACAS has positive operating income and is set to grow that operating income, the case for maintaining a NAV discount seems weak. The overhanging bankruptcy scare is long gone, and the future looks bright.

What kinds of opportunities does ACAS have to increase NAV? For an obvious example, let's look at ECAS – the European arm of the business that ACAS took private in early 2009. The NAV of ECAS is $0.7B, but ACAS claims it has a FAS-157-compliant "fair value" of $0.4B, or a $0.3B discount. The $0.4B given under FAS 157 is again discounted in the hands of ACAS buyers, who get their share of ECAS at a 42% (or so) discount, valuing ACAS investors' ECAS ownership at a total of $0.23B (or thereabouts). Would you buy ECAS for $0.23B when its NAV is $0.7B? What if you heard some of its portfolio companies were being given harsh discounts under FAS 157 and were worth a lot more than appeared on the books? Considering that ACAS has spent most of its publicly-traded life trading at a premium to NAV, the prospects for a multiplying effect as NAV discounts evaporate is an exciting and plausible prospect.

With respect to buying mispriced assets, ACAS founder Malon Wilkus said this:

We are now focused on originating high quality investment opportunities while continuing to improve our balance sheet. We have generated a 27% return on equity since the second quarter of 2009 and believe we will continue to generate strong book value growth for our shareholders as we recover from the recession.

The President of ACAS' Specialty Finance and Operations, Gordon O'Brien, spoke more specifically about this:

The quality of the portfolio continues to improve. Our investment and operations teams have made good progress in improving many of our troubled companies and the pace of new troubles developing in the portfolio has slowed significantly. We are pleased to now be able to turn our focus to making new investments. History has shown that the most attractive investment opportunities are made during the recovery from a recession and we believe that will continue to hold true.

ACAS expects to hunt down the most attractive investment opportunities exposed by the economic chaos and the recession, and use cash to pluck illiquid and discounted middle-market companies from the hands of exiting owners for the benefit of shareholders. This is exactly the reason I liked ACAS in the first place: it gets to cherry-pick the best deals in a field thin with buyers and flush with sellers.

Keeping Up With the Jonses
Normally, I don't spend much time worrying about what other people are doing with their investments (except to the extent I think they are creating an opportunity, like overselling ACAS or underestimating AAPL). However, there seem to be quite a few funds betting on the rationalization of ACAS' share price. ACAS maintains a list of major stockholders, and quite a few funds seem to hold a tenth of a percent of ACAS' outstanding shares or more as of their most recent filings. Among them are Paulson & Co. (which bought in a private placement in April), Vanguard Group, Dimensional Fund Advisors, and BlackRock – each of which reportedly holds over 3% of ACAS' outstanding shares. Well, in Paulson's case, over 12% of ACAS' shares outstanding. These are some big, ten-million-plus share bets.

They aren't making these bets because it looks like a crapshoot or because the anticipated return looks lame. I think they're recognizing that buying ACAS under these conditions is shooting fish in a barrel, or fishing with explosives: you can't hardly miss.

===
[1] Eventually, ACAS will have enough assets that will look to refinance its debt because it wants terms that improve its ability to leverage up when the investment cycle suggests that leveraging up makes sense. That will be a few years off, though. For now, we have lowered interest expense, lowered need to lobby lenders not to exercise acceleration and related default rights, and lowered need to reimburse lenders for legal expenses in connection with debt enforcement under oppressive debt agreements that force borrowers to bear lenders' legal expenses. (You laugh; as the fireman said in Ghostbusters, I have seen shit that would turn you white. I'm betting some of that $17m was reimbursement of lenders' legal expenses.)