PDF version of report with charts and tables:

https://docs.google.com/file/d/0B86krX6VxmmWd0R6b0tZZy1uLUU/edit?usp=sharing

Short a pile of assets trading at an EV / EBITDA that exceeds the remaining life of its assets as the fundamentals deteriorate.

Short TAL.

TAL trades at 8.7x run-rate EBITDA despite having assets with a remaining life of 7 years and a significant rent-roll problem. It trades at a 100%+ premium to replacement value, and we believe EBITDA will fall in 2013, missing analyst estimates by a wide margin. That's probably why the PE sponsor hit the bid a day after the 10K was filed in February.

Ticker: TAL

Stock Price: $45

M Cap / EV: $1.5B / $4.0B

Debt: $2.64B/2.54B gross/net debt (5.65x gross leverage; 5.2x net leverage using standard EBITDA)

Avg Daily Volume:  352K shares (~$15.8m)

Valuation:

EBITDA: 8.7x FY12; 8.2x FY13 (consensus); 7.9x FY12, 7.6x FY13 (adding back Gains on Sale)

Historical range: 6.0-7.5x (street includes GoS); ; Street typically anticipates , generally trending lower

Dividend Yield: 5.7%

P/Tang Book: 2.25x (1-3.25x range; peers in-line)

Base Case: $25 -- 6.5x 2013 EBITDA (incl GoS); Approx replacement value ex DTL.  

Downside/Risk Case: $46 -- 8.0x 2012 EBITDA (incl GoS)

Analyst Recommendations:  5 Buys, 4 Holds, 0 Sells

Notable Holders:  Resolute/Jordan 0.0% (see secondaries below), mostly vanilla MF holders.

Insiders own 2.0%, 1/4 of which is associated with restricted stock (2 directors elected for cash in lieu of stock)

Exec comp criteria is based on Adj EPS target (target % of salary as bonus is 65%, max is 130%; hit 100% in 2012 with benefit of depreciation change; 123% in 2011).

CEO owns 307k shares, inclusive of 75k restricted stock

Secondaries:

March 2011: $36 / 5.5m shares - 2.5m primary, rest was PE sponsors

March 2012: $36.42 / 3.0m shares - all secondary from PE sponsors

May 2012: $38.50 / 5.0m shares - all secondary from PE sponsors

Feb 2013: $42.90 / 4.0m shares - all secondary from PE sponsors

Thesis:

1. Deteriorating container market fundamentals - rates and utilization rolling over
2. Gains on Sale included in EBITDA are misleading and set to decline
3. Change in depreciation policy allows for "double dipping" with Gains on Sale
4. Significant insider selling
5. Aggressive adjustments/disclosure modifications around swaps, leases, and utilization.
6. Substantial downside in recession, including risk of insolvency as DTL unwinds at worst possible time.
7. Peak valuation

Business Overview

TAL International Group, Inc. owns containers and leases them out to global liners on mostly long-term contracts (5 year initial term on new assets). It primarily leases three types of containers Dry freight containers, refrigerated containers and special containers. The dry freight containers are used for general cargo, refrigerated containers are used for perishable items, and special containers are used for heavy and oversized cargo.

The company was created when the Jordan Companies purchased Aegon's container leasing business after a trying period in the container market.

The industry is comprised of global liners (Maersk, CMA CGM, COSCO, etc), containership lessors (Seaspan, Danaos, etc), and container lessors (TAL, Textainer, etc) on the ocean going container front. Terminals, rails and freight forwarding companies also factor into the chain, but they aren't particularly relevant to this analysis. Global trade is obviously the main driver of the business, and growth of containerized trade has outpaced global trade growth as more goods are shipped in containers.

The global liners own ships and containers, but they also lease them to manage capital/capacity. Drewry Maritime estimates that 45% of the worldwide fleet is owned by lessors. This is up from 42% in 2006 (when TAL noted that the % had ranged from 42-47% over the last decade). TAL and its leasing peers are small, commoditized suppliers to the larger, more concentrated liners. Clearly liners are going to use their own capacity first and allow leases to roll-off when not needed. With that said, having the infrastructure and depots to manage the inventory as it comes off lease for re-lease and sale provides some benefits to scale. Additionally, the industry has moved toward longer lease terms with ~70% of the market being on leases of 4-5+ years versus less than 50% around 2000.

How did we get here?

While TAL was facing a terrible situation during the 2008/9 downturn (liners on the cusp of BK, credit/refi concerns, and facing the risk of a DTL unwind without new asset purchases), they ultimately came out big winners.

Demand fell rapidly as retailers and manufacturers worked off inventory, but demand came back relatively with monetary/fiscal stimulus.

At the same time, container manufacturing ground to a halt. Two Chinese manufacturers control 70% of the market, shut down production (note utilization at ~10% for 2009 in the chart below) and sent workers home.

So supply fell as old assets were retired and capacity did not come back on-line quickly with the recovery. Additionally, effective supply shrank ~4-8% because of slow steaming. The liners weren't being compensated for rushing, so they cut costs by slowing the speed of ships, creating a need for more ships/containers to deliver the same amount of goods each year.

Good background piece:

http://www.worldshipping.org/public-statements/2011_Container_Supply_Review_Final.pdf

Key Issues:

1. 1.       Deteriorating Fundamentals - rates are falling

Below is inudstry data for new 5 year leases:

We are starting to see this show up in TAL's effective per diems (rates can vary from the above due to utilization, timing of lease initiation, etc ), and this data and 10K commentary corroborates the industry data directionally:

TAL 2012 10K:

                "market lease rates for dry containers are currently below our portfolio average"

This makes sense because lease businesses are commoditized and should only earn their cost of capital over the course of a cycle. Container lessors are swing capacity to the global liners, and liners have been stressed for a prolonged period and have needed liquidity, so container lessors have been able to earn excess returns. But the situation is untenable.

Below are the key drivers of different aspects of the container market. In the end, long term rates cannot remain high with stable/falling steel prices. In a recent call with TAL, the reiterated that lease rates are very closely tied to newbuild pricing which is mostly driven by steel prices (60-70% of COGS).

The data and graph below is pulled from TGH data on ASP/residual values and HRC data available on Bloomberg. The correlation coefficeint is 0.47 and the R-squared is 0.60. Given where steel prices are today, I think 2005-2007 is a decent proxy for where we could go in terms of newbuild pricing and residual values.

As mentioned earlier, the box manufacturers shut down in 2009. They were slow to ramp in 2010, but have subsequently ramped production and added capacity (ie: Singamas).  Now that they have caught up to (overshot) demand and steel prices are falling, newbuild prices could fall further and result in even more pressure on rates.

Below are estimated returns for a standard 20' dry container based on Cointainerisation International's data (obviously ignores costs). Note the significant decline in newbuild price and per diem rates.

1. 2.       Deteriorating Fundamentals - Utilization is under pressure

Utilization is also falling, but you woudln't know it by looking at the company reported data (largely because of two significant changes in definition in recent years):

The table below uses the company's footnote disclosures to back into a utilization proxy:

1. 3.       Change in depreciation policy

The company changed its depreciation policy as of 10/1/2010 and again at 10/1/2012 to increase the residual value.

I'm not contending that the residual value is wrong or that the accounting is wrong, just that the change will only have a short term impact on earnings before becoming a drag on earnings/EBITDA (because they include GoS in EBITDA) -- initially a boosting earnings, then causing earnings to revert as gains on sale are lower (because of the higher basis). In effect, the company was over-depreciating in the past, and then recapturing the excess depreciation in gains on sale (though if they hadn't over-depreciated, it never would have hit EBITDA). Until this change in accounting assumptions works its way through, the company will be "double-dipping" on earnings.

Below are the actual changes:

The changes affect different aged assets differently:

• Assets that have already been depreciated to below the new residual value will no longer incur depreciation charges.
• Assets that are partially depreciated but above the new residual value face less depreciation per period because the remaining total depreciation (to the new residual) is spread over the remaining asset life.
• New assets will depreciate less quickly (20' dry below):

The company has undergone a fleet renewal in recent years, resulting in a lower number of units that are fully depreciated. A younger fleet is generally a positive thing, but it will be a negative for earnings as more units are subject to depreciation (to the extent investors are focused on earnings), and it will be a headwind to earnings/ebitda in future years as gains on sale diminish (higher basis).

1. 4.       Gains on sale will decline

This issue is related to the one directly above, but is probably more impactful and less appreciated.

The market for used containers has been consistently better than prior residual assumptions (largely a function of rising steel prices over that period) resulting in consistent gains on sale of used assets. From TAL June 2012 investor day (showing used pricing indexed to 4q06):

The inclusion of gains on sale in EBITDA is misleading because it's really just a re-capture of historical over-depreciation, as discussed above. Ideally you start with the right assumptions and don't make any adjustments. Next best would be consistency of policy and netting GoS off of depreciation. The change in policy skews things for earnings and putting it in EBITDA on top of that is a little egregious. With that said, the policy isn't likely to change, so we have focused on what is likely to happen to reported results.

The gains per unit will decline as (1) the market normalizes (the company does too, see below slide) and (2) the higher residual assumption discussed above starts to flow through.

The effect can already be seen to a degree. In the table below one can see that margins on the sale of used containers is down from peak levels (and still falling), and despite the company selling more units (positive delta from volumes), the company had a decline in GoS in 2012.

There will be further pressure on this line from the limited supply the company will have of assets that are hitting the end of their life. This is because they made limited investments in 1997-2003 when they were part of larger finance companies (TransAmerican, then Aegon) and the container market was weak.

My numbers are approximations based on comments in filings. I've treated TEU purchases as newbuilds which will understate the number available for re-sale in the coming years, but this is how the company suggested modeling it. Even considering the sale-leasebacks they've done, they're unlikely to have significant assets available for sale because the company typically buys assets with at least 7 years of remaining life (per IR) and has done the majority of its buying in 2010 or later.

Some comments to support this assertion:

2008 10K:

In 2008 we sold approximately 78,000 TEU of our older containers, excluding containers purchased for resale. Over the last three years, our disposal of older containers has averaged approximately 70,000 TEU per year, which represents an average of approximately 7% of our equipment leasing fleet at the beginning of each year. This 7% annual disposal rate is close to our theoretical steady-state disposal rate given the 12-14 year expected useful life of our containers. However, based on the age profile of our existing fleet, scheduled lease expirations and the prospects for decreased leasing demand due to reduced trade growth, we expect our rate of disposals will increase in 2009 and remain elevated before decreasing several years thereafter. A disproportionately large share of our containers were produced before 1997 since we invested in a relatively small number of containers from 1997 through 2003. We expect that we will sell most of the pre-1997 units over the next several years. During years of above-average disposals, our TEU growth rate may be constrained if we are unable to generate a sufficient number of attractive lease transactions for an expanded level of new container purchases.

2009 10K:

                The age of our container fleet may become a competitive disadvantage.

As of December 31, 2009, the average age of the containers in our fleet was 7.5 years, and the average age of our fleet increased by 0.4 years in 2009 due to our limited procurement of new equipment. We believe that the average age of most of our competitors' container fleets is lower than the average age of our fleet, and customers generally have a preference for younger containers.

...

The age of our fleet may result in an increase in disposals of equipment and result in a reduction of lease revenues if we are unable to purchase and lease similar volumes of equipment.

As of December 31, 2009, the average age of the containers in our fleet was 7.5 years. A large portion of our fleet was acquired in the mid-1990's and is on leases which take the units to the end of their serviceable life in marine transport. Upon redelivery, this equipment is likely to be disposed. From 2005 through 2009, we sold on average approximately 7% of our equipment leasing fleet containers annually. However, the rate of disposals in 2009 did not keep pace with the rate at which older containers were returned to us, and our disposal rate would have been close to 10% in 2009 if the sale market had been strong enough to allow our sales to keep pace with returns. Due to the significant portion of older containers in our fleet and the increase in our inventory of containers available for sale we expect that our disposal rate will increase in 2010 and remain at a historically high level for several years thereafter.

2012 10K:

Equipment disposals. During 2012, we recognized a $44.5 million gain on the sale of our used containers compared to $52.0 million in 2011. Gain on sale decreased primarily due to lower average sale prices and the higher cost of equipment sold, partially offset by higher sales volumes. The cost of equipment sold in 2012 was higher than in 2011 due to the substantial number of sale-leaseback containers with higher book values sold during the year. Those sale-leaseback containers had been purchased in the latter part of 2011 and second half of 2012 for prices higher than the typical book value of our older containers.

1. 5.       Misleading disclosure / Abuse of Non-GAAP adjustments
   1. a.      Exclusion of swap gain/loss

This is a relatively minor issue, but they exclude the gain/loss on swaps from their adjusted net income. That's fine because mark-to-market overstates the impact, but it also understates their interest expense (~60% of debt is floating rate and swapped, but the hedge doesn't get hedge accounting treatment so it goes through at Libor plus a small spread even though they're paying more out).

1. b.      Including gross finance lease principal payment  in adjusted EBITDA

These are essentially reverse capital leases where they rent out their balance sheet to lessees. TAL generally doesn't bear residual risk (other companies in the space note bargain purchase options that result in limited/no residual risk).

To account for these leases, the company puts the net finance receivable on the balance sheet. This is comprised of a gross finance recievable (subject to ADA), net of an unearned liability portion that is the imputed interest to be earned over the life of the lease. Including the income portion (which is accreting through the income statement) in EBITDA is debatable because it's very much like owning a bond/interest income (i would net it off of debt). However, my big contention is with the fact that they add back the principal payment associated with these leases to EBITDA to get to "Adjusted EBITDA". Yes, they are receiving a cash payment, but that would be like considering the entire proceeds on sale of an asset as income rather than just the the gain.

This is something of a qualitative point because their disclosure is is not likely to change, so I don't want to over-emphasize it. However, I think it is another example of TAL going out of their way to make the numbers better than they are.

1. c.       Misleading utilization definition (and multiple changes to the definition)

I think the company overstates the steadiness of the business by manipulating utilization. First, they exclude assets bought that are not yet on lease--essentially excluding spec purchases from utilization. If utilization disclosure isn't for that purpose, i'm not sure it has a purpose. They can also designate old assets as "for sale" to avoid hurting utilization (part of the recent change). This is very grey because an asset might be sale-able at 12 years, or last to 14+ years. Additionally, they could slide things in/out of this category depending on lease demand.

Second, they changed the way they calculate utilization to focus on CEU and exclude assets designated as "for sale" (discussed above). The shift to CEU is important because dry containers are the most volatile (handle manufactured goods v. refrigerated that handle food and are higher CEU/TEU). This is not a smoking gun, but it is yet another red flag.

Utilization. Our average utilization was 98.7% for the full year of 2011, an increase of 1.1% from the average for 2010. Ending utilization decreased 0.1% from 98.7% as of December 31, 2010 to 98.6% as of December 31, 2011. Our dry container utilization was exceptionally high throughout 2011, even as the pace of new container lease-outs slowed in the second half, due to the general tight supply/demand balance for containers, and as shipping lines retained existing leased containers on-hire in 2011 in anticipation of strong peak-season demand for containers, and to benefit from low historical lease rates. However, container redeliveries began to return toward historically normal levels toward the end of the year, after the end of the peak season for dry containers, and our utilization decreased slightly in the fourth quarter.

Utilization and leasing demand for our refrigerated containers remained strong in 2011. Leasing demand for special containers remained steady while leasing demand for tank containers was strong. Utilization for our chassis product line improved due to a better supply/demand balance for chassis in the United States.

The following tables set forth our equipment fleet utilization(1) for the periods indicated below:

(1)

Utilization is computed by dividing our total units on lease (in CEU's) by the total units in our fleet (in CEU's) excluding new units not yet leased and off-hire units designated for sale.

Effective for this filing, we have changed our utilization calculation to be based on CEU's and to exclude off-hire units designated for sale. This new method provides a better indicator of the performance of our leasable fleet because it gives greater weight to more expensive equipment types and it does not include those off-hire containers that have been designated for sale. In addition, we believe our utilization calculated under this new methodology more closely conforms to those used by our publicly traded competitors. Utilization for all of the periods shown above has been recalculated using this new methodology. Utilization under the former methodology would have been 97.4% as of December 31, 2011, and the average for the quarter and year ended December 31, 2011 would have been 97.9% and 98.3%, respectively.

1. 6.       Deferred Tax Liability

The company uses an accelerated depreciation policy for tax purposes. GAAP is straight-line to a residual over 12-14 years, but the company is able to depreciate the assets (whether new or used) to a zero residual over five years for tax purposes (per IR).  This allows them to shield income with the additional depreciation when they are purchasing assets. However, it will force them to keep growing in order to avoid unwinding the DTL (as they hit depreciation cliffs from large investment years). This looked to be a significant concern in 2009 when there was no demand for containers and they faced the prospect of paying a significant amount of cash taxes as their growth push from 2004-2008 was about to start hitting depreciation cliffs.

With the rebound in trade and container demand, the company was able to avoid this outcome. It likely happens someday, but they’ve adequately deferred it for now by purchasing a significant amount of assets in recent years. The amount isn’t overwhelming in good times (388.5m gross DTL, 270m net), but it would only come due when they could least handle it – a time of very weak container demand where the secondary market is also weak. As long as there is any container demand, they’ll be incentivized to grow assets – and keep growing them aggressively enough to shield income regardless of expected return unless they don't have the liquidity to do so.

1. 7.       Credit quality of customers

The global liners have been reporting poor results since the downturn and CMA CGM has been a contsant restructuring story. This was not a meaningful issue in recent years because existing leases were below market. However, lessees are likely to negotiate more intensely given the recent declines in rates, and the fact that existing leases are now above market rates (see quote above in rates section).

1. 8.       Risk associated with spec orders not leased up yet.

As discussed earlier, the company's utilization is overstated as a result of an aggressive definition. In addition to the equipment listed below, TAL has $73m of equipment purchase commitments that have not yet been completed. This provides them a shelf of inventory to lease out, but it also makes them more desperate to sign bad leases given the risk of further redeliveries as leases expire (especially on high rate leases).

The street gets excited about asset growth here because it is short term accretive, but i'd be surprised if capex was meaningfully higher than ytd (inclusive of the purchase commitments) until 3q. This could come as a disappointment and cause revenue/EBITDA to fall short of expectations.

Valuation:

TAL has only traded higher than it does today for brief periods when rates were rapidly rising and well above their existing lease rates (positive rent roll). We have the opposite situation today.

Catalysts:

• Missing Rev/EBITDA/EPS estimates from:
• Container market softening - falling utilization and rates working into #s
• Additional expenses associated with idle equipment
• Fewer gains on sale / lower gains per unit
• Retiring old units and deploying the capital into new units (EBITDA per capex spend will be lower)
• Potential for DTL unwind (not near term)
• Restructuring of customers
• Difficulty leasing spec orders

Risks:

• Low P/E; not expensive on EBITDA either.
• Slight discount to TGH (but more aggressive accounting/insider selling here)
• Potential sale to strategic (liner co) especially if container market remains tight (long shot)
• Dividend appears secure and is relatively high at 5.7%
• Accretive sale-leaseback opportunities

• Missing Rev/EBITDA/EPS estimates from:
• Container market softening - falling utilization and rates working into #s
• Additional expenses associated with idle equipment
• Fewer gains on sale / lower gains per unit
• Retiring old units and deploying the capital into new units (EBITDA per capex spend will be lower)
• Potential for DTL unwind (not near term)
• Restructuring of customers
• Difficulty leasing spec orders