This section sets out the sources used in conducting the research on First Allied Corporation so that interested readers can examine the original materials themselves. I am also keen to demonstrate that the information published by this blog is sourced from publicly available information from First Allied's lenders, legal documents filed by the company with government agencies and from First Allied itself. Where calculations or projections have been made, I set out the methodology used to ensure the research can be fully understood by readers.
The process of collecting information on First Allied begins with county and state records which show which CMBS contains the mortgage for each centre. Government data is also used to determine, where possible, when exactly the property was acquired, for what price and whether First Allied has previously taken out mortgages on the centre. CMBS filings then provide details of the terms of each mortgage and the current financial performance of the assets. This financial data is brought up to date using the latest occupancy information from First Allied itself.
CMBS 425B5 filings
All US CMBS vehicles must file a prospectus with the Securities and Exchange Commission (SEC). The final version of the prospectus must be filed in accordance with rule 425(b)(5) of the Securities Exchange Act 1933 and thus such prospectuses are often called 425B5 documents.
The 425B5 documents are very long (six to seven hundred pages) and give a lot of information on the functioning of the CMBS itself, but also provide a significant level of detail about the properties which secure the CMBS' pool of mortgages. They are a major source of information on First Allied's properties at the time their mortgage was taken out. Information derived from 425B5s includes:
| Item | Comment |
| Property name | |
| Control number | The unique number given to each property (these no.s, rather than names, are used to identify properties in CMBS reports) |
| Property address | |
| Original mortgage balance | On the date the mortgage was taken out |
| Cut-off date mortgage balance | Balance on "cut-off" date when CMBS constructed, for interest only mortgages this will be the same as the original balance |
| Mortgage rate | All First Allied mortgages are fixed rate |
| Amortization type | Is the mortgage: interest only / repaid equally over its life or after an interest only period / is there a "balloon" payment at its end |
| Length of any interest only period | |
| Mortgage term | Length of mortgage in month |
| Amortization term | In months. Balloon or "ARD" mortgages will amortize (i.e. be repaid) as if they were longer than they actually are, leaving an outstanding balance at the end |
| Origination date | Date mortgage taken out |
| Maturity or anticipated repayment date | Date the mortgage ends or (in the case of an ARD loan) is anticipated to end |
| Balloon amount | Where relevant, the amount of any balloon payment at the end of the mortgage |
| Annual debt service cost | The cost of interest and (where appropriate repayment) costs at the start of the mortgage |
| Underwritten net operating income | The property's net operating income (NOI) at the time the mortgage is taken out |
| Underwritten net cash flow | The property's net cash flow (NCF) at the time the mortgage is taken out |
| Underwritten net cash flow DSCR | The ratio by which debt service costs are covered by net cash flow at the time the mortgage is taken out |
| Appraised value | The independent value of the property undertaken for the original lender |
| Appraisal date | |
| LTV | Loan to value at the at "cut-off" date |
| Year built | The year the property was constructed |
| Year renovated | The year(s) of any major renovations |
| Square footage | The square footage of the property |
| Occupancy | The percentage of the property let at the "cut-off" date |
| Names of three largest tenants | |
| Space occupied by three largest tenants | |
| Expiry dates of three largest tenants' leases |
I have extracted the relevant information from a 455B5 for one example centre, McDermott Commons, to give an example of the format in which the information is shown. A document showing the extracts can be viewed downloaded here.
The following tables give links to all of the relevant 425B5 documents containing details of current mortgages outstanding on First Allied's shopping centres as well as mortgages on those centres that have gone bust or been sold. The list of which centres' mortgages are in which CMBS is too long to show in this section, but can be viewed here.
In analysing First Allied, the 425B5 documents give a very significant amount of information about each property, including crucially First Allied's equity in each property at the time they were acquired or remortgaged (the appraisal value less the balance on the mortgage) as well as significant details on the shopping centres' financial performance in terms of NOI, NCF, and occupancy rates.
Where a centre has been in two (or more) CMBSs over time, a very clear picture of the changes in the value of the property and the amount of equity withdrawn by remortgaging can be gained by comparing loan value and appraisal values in the different 425B5s.
The 425B5s disclose whether the relevant property has additional mortgage debt secured on it but which is not included in the CMBS. This applies to 36 of the centres, all but four with Lehman Brothers mortgages and all with mortgages dating from 2004, 2005 and 2006. In these mortgages, the lender splits the loan into two "notes", "A" and "B". The notes are "cross collateralised" which means that defaulting on payments on one means the borrower has defaulted on both. The A notes ranks above the B note in cases where the security has to be realised. The purpose of splitting such mortgages in this way was to "create" loans suitable to be sold into various markets. The A/B split could be done in such a way as to ensure the loan to value of the A note made it suitable to be placed in a particular CMBS. The B note (which receives far higher interest than the A note, typically 12.5%+ vs. 5-7%) would be sold on by the lending bank to hedge funds seeking high yield, higher risk securities. This A/B structure made no difference to the borrower who had to keep up combined payments at a competitive blended rate on the two notes. A/B loans are a classic credit bubble example where (even though the B note is disclosed) a structured product (in this case the CMBS) is dressed up to look lower risk than it is.
The other information that is of great importance is the structure of the loans on each centre. In particular, the length of interest only periods for those mortgages which have them is key in understanding what will happen to the financial performance of the First Allied portfolio. During the credit boom, the terms on which commercial mortgages could be secured became ever better for borrowers allowing greater leverage. Properties were often acquired using very high LTVs, where (just as with UK and US housing loans) the loans were made "affordable" by the presence of interest only periods during which no actual repayments were made. The idea was for the borrower to refinance into loans with new interest only periods or to sell the property before these periods ended. The credit crunch and ongoing lack of mortgage financing has made such new refinancing very difficult (and of course impossible for properties where the fall in value has pushed them into or close to negative equity). Many borrowers are thus facing the prospect of actually making repayments on loans where they never expected to. If the property's income and cash flow generation has fallen because of the recession, loans coming off interest only periods can become unaffordable.
Under section 15 of the US Securities Exchange Act 1934, CMBS vehicles (because their securities are not listed on an exchange but instead are traded "over the counter") do not have an ongoing obligation to file reports with the SEC. It would however be impossible to sell CMBS securities without making a commitment to potential investors to provide ongoing reports about the performance of the properties securing the pool of mortgages and the trustee of each CMBS provides a series of monthly reports. These are pseudo public documents requiring registration on the trustee's website. After a series of mergers, the market for trustee services has consolidated down to four players of which three act as trustees for CMBS containing First Allied's loans (or loans on properties that have gone bust). The three companies and their websites are:
| Trustee company | Website address |
| Bank of America Merrill Lynch | |
| Wells Fargo | |
| Bank of New York Mellon |
Borrowers have obligations under their mortgage agreements to report on the performance of their properties on a quarterly basis and mortgage payments are made (or not made) monthly. The trustee of each CMBS publishes a monthly pdf report showing summary statistical information on the performance of the CMBS itself and information on each underlying property. In addition to these reports (see table below for links to the most recent versions), the trustee also makes available spreadsheets containing far more detailed information about the underlying properties. These spreadsheets effectively update much of the information in the original 425B5 documents and (in most but not all cases) actually go further by giving full income and cashflow statements for each property (see example below). This information is also shown in a comparative format against the performance data of the property when the mortgage was taken out and over each of the last two years. The reports and/or spreadsheet also give commentary on those properties that have been placed on "watchlist" as potential problems as well as those which loans which have become delinquent and have thus been passed onto the "special servicer".
The ongoing filings give quite up to date information on rental and other income, costs and capital spending at each of First Allied's centres and also show what the level of occupancy was at the reporting date, we can quite easily work out what each level of NOI and NCF each centre is currently generating at today's occupancy rates. The filings allow us to see, in extraordinary detail, how much income and cash flow each of First Allied's centres is generating and to compare this to the "Principal and Interest" (P&I) cost of its mortgage. If a large tenant leaves a centre, we can calculate to a very high level of confidence what impact this will have on the asset's profitability and ability to pay its mortgage.
The vast trustee spreadsheets obviously include information on thousands of other properties as well as First Allied's. To make this information easier to access, I have consolidated all the relevant information on the financial performance of First Allied's shopping centres into one spreadsheet. It can be downloaded here.
I have consolidated the trustees' commentaries on the 28 of First Allied's centre that have been placed on "watchlist" into one sheet. It can be downloaded here.
First Allied's website and loopnet.com
As a property company, First Allied obviously wants to fill its vacant space and needs to advertise it. This makes the company's own website (http://www.firstalliedcorp.com/) an important resource. The company lists each centre it currently owns (properties that go bust are obviously swiftly removed but cached copies will be on the web forever) and show maps and details of every property including which lots are vacant. Because First Allied use the site as a shop window to advertise vacant property to potential tenants, the information is far more up to date than that reported (quarterly with several months' lag) to the CMBS trustees. I have used the occupancy rates shown on the company's website in my analysis of what level of NOI and NCF the portfolio is currently generating.
The company's website (and cached copies of it) also has a number of press release relating to the purchase of centres over the years. These press releases generally state the price paid and the month and year in which the centre was acquired. There is also normally a bullish comment from Edward Glazer about how great the centre is!
Althought First Allied's website shows which lots in which centres are vacant, the properties are actually advertised, with asking rents, on the huge commercial real estate website http://www.loopnet.com/. This information is very valuable in ascertaining how the rents the company is seeking compare to what is being received by current tenants at the same centre (which we can see from the CMBS trustees' spreadsheets). In an environment of falling rents and a weak economy, even commercial properties that manage to fill vacancies can end up in big trouble if they have to accept very low rents (or rent free periods or other concessions) to do so.
Online county and state records
Under US law, borrowers must file certain real estate records. All real estate deeds (sale and purchase contracts and mortgages) must be filed with the county in which the property is located and are almost always available for public viewing (there are privacy restrictions in a minority of states). Counties raise most of their tax revenue from real estate taxes with properties appraised annually. These appraisals and other supporting information, including ownership and sale histories are publically available from the relevant county appraiser. Under Federal law, "Uniform Commercial Code" (UCC) documents describing business assets used as security for loans must be filed with the relevant state.
Most states and counties allow online access to all these public records. A few counties have either not made records available online or have only published very limited details. UCC filings for Delaware companies can be accessed, but only by using (expensive) approved Delaware search providers. For the great majority of First Allied's properties, I have been able to use UCC and/or county records and appraisals to determine the ownership history of each property, the price originally paid, the First Allied company that owns the property and the lender (and subsequent CMBS) for each mortgage.
A full list of each centre, its mortgage, financial performance and which state and county can be found on the master spreadsheet. Occupancy data can be downloaded in a spreadsheet too. Examples of tax appraisal records and full mortgage deeds are available on request.
Pro forma NCF and DSCR figures
As described above, each property reports financial data to its CMBS trustee quarterly with a lag of a few months. In the May trustee reports, two of First Allied's sixty three centres with securitised mortgages had only reported financial data to the end of 2008, nine had reported information for the period 1 January to 30 September 2009 whilst the other 52 had reported financials up to 31
December 2009. The First Allied website shows the current level of occupancy at each centre. Where the occupancy rate at a property has risen or fallen by five percentage points or more since the last reported date, I have calculated "pro forma" net cash flow and DSCR figures using the level of occupancy shown on First Allied's website.
December 2009. The First Allied website shows the current level of occupancy at each centre. Where the occupancy rate at a property has risen or fallen by five percentage points or more since the last reported date, I have calculated "pro forma" net cash flow and DSCR figures using the level of occupancy shown on First Allied's website.
For the twenty eight centres whose interest only mortgage periods end in 2010, I have calculated pro forma year one principal and interest charges and DSCRs to show the impact of repayments starting.
The table below shows how the pro forma figures are derived for a centre (Ulster Terrace in Colorado) which has only reported financials up to 30 September, has seen a major change in occupancy since that date (+8.6%) and is coming off an interest only period in 2010:
Item | Comment | |||
| CMBS | A | LBUBS 2005-C5 | ||
| Last reporting date | B | 30-Sep-09 | From May 2010 trustee report | |
| Interest only period ends | C | Apr-10 | From 425B5 | |
| Interest rate | D | 5.71% | From 425B5 | |
| Occupancy at reporting date | E | 70.1% | From May 2010 trustee report | |
| Current occupancy rate | F | 78.7% | Latest figure from First Allied website | |
| Current rate as % reported rate | G | 112.2% | F / E | |
| Financials for period 1 Jan to 30 Sep 09 | Annualised | |||
| Income | H | 647,557 | From trustee NOI report. Reported no. / 0.75 to annualise. | |
| Costs (inc capital costs) | I | 375,734 | From trustee NOI report. Reported no. / 0.75 to annualise. | |
| Net cash flow | J | 271,924 | H – I | |
| Interest payments | K | 364,461 | From trustee NOI report. Reported no. / 0.75 to annualise. | |
| DSCR | L | 0.75 | J / L | |
| Proforma financials at current occupancy | Annualised | |||
| Income | M | 726,943 | H x G. Assume space is vacated/taken at average rents. | |
| Costs (inc capital costs) | N | 375,734 | Held flat at annualised reported level. | |
| Net cash flow | O | 351,209 | M - O | |
| Interest payments | P | 364,461 | Held flat at annualised reported level. | |
| DSCR | Q | 0.96 | O / P | |
| Proforma P&I costs after expiry of IO period | Annualised | |||
| Amortization length in months | R | 360 | From 425B5 | |
| Current principal balance | S | 6,565,000 | From April 2010 trustee report | |
| Annual repayment | T | 218,833 | (S / R) x 12 | |
| Principal balance Y+1 | U | 6,455,583 | S - T | |
| Interest cost on average balance Y1 | V | 368,614 | D x (average of S and U) | |
| B Note P&I payment | W | 50,368 | From Trustee NOI report | |
| Total P&I cost | X | 637,815 | T + W | |
| Pro forma DSCR | Y | 0.55 | O / X |
The two major assumptions which some readers may disagree with (both made largely to ensure transparency), are the assumption that income falls or rises at the same rent per square foot as prevails in the rest of the centre and the assumption that costs are fixed (at least in the short-term). The first assumption probably exaggerates net cash flow where occupancy is rising as rents on newly let sites are likely to be lower than passing rents in the centres. I am happy that examination of the detailed net cash flow statements will reassure readers that the second assumption, that costs are largely fixed, is sensible. The major costs at First Allied's centres are property taxes, repair and maintenance and general and administrative costs. I believe that all these will need to be continued to be paid in centres with falling occupancy. Repair and maintenance expenditure may theoretically be variable, but in reality vacant lots are unlikely to be filled in badly maintained centres.
Valuation estimates
After some debate, I have decided to use a simple but completely transparent methodology for making an approximate estimate of the gross asset value (i.e. before deducting debt) for each property. Using data from loopnet.com, I calculated that neighbourhood centres in Atlanta, Dallas and greater Washington DC are being offered for sale on an average (weighted by the offering price) capitalisation rate of 8.5-9%. As I am deliberately using net cash flow, I decided to use the lower end of this band and for simplicity, and to apply it across the portfolio.
Clearly, this is a simplification. Each property is different and each local and regional real estate market is different. I do not profess to be a US real estate expert, far from it, and am not qualified to pronounce on the "correct" valuation beyond such a simplistic assumption. What encourages me that using a single cap rate produces reasonably meaningful results is that cap rates on centres on the market tend to be very bunched around a range of 7.5% to 10.5%. Even the First Allied centres that have gone bust (which by definition had very low occupancy and hence low cash flow) where one would expect to see low cap rates (to reflect their recovery potential) have been offered on rates in this range (Crosswoods Commons is advertised at $1.5m or a cap rate of 8% and The Shops at Cumberland Place at $3m or a cap rate of 7.8%).
