Description

I am recommending HOG as a short because its captive finance subsidiary is misunderstood and will generate huge negative surprises, and because the core motorcycle business is fully valued on "recovery" earnings in 2011.  Negative earnings revisions in the next one or two quarters, and a likely retreat from the current "full" valuation, which does not price-in any risk to a recovery, should cause the shares to drop 20-40% back to levels before its mid-summer short squeeze.

HOG is two businesses, the motorcycle maker (Motor Co.) and a finance sub (HDFS, short for Harley Davidson Financial Services).  The sell side is assuming HDFS will be slightly profitable going forward although I think it will have significant losses.  The sell side essentially ignores this business because they are not bank analysts, it has always been a very small contributor (i.e. <10%) to earnings, it gets consolidated into two lines on the income statement, and the accounting in complex.  To be clear, the finance sub had a liquidity crisis earlier in the year, but they have largely resolved any liquidity problems, so this is not part of my short thesis. 

The key issue is that HDFS has historically sold most of its loans off-balance sheet via securitizations, but due to an accounting change at the end of last quarter, it is now holding all loans on balance sheet and it will have to account for credit losses like a bank.  In other words, it will have to take large provision-for-loan-loss expenses every quarter to maintain an allowance for loan losses on the balance sheet.  HDFS has always had a provision expense for the small portion of loans it has kept on balance sheet, but it is now going to have to provision for the $2.7 bn of loans it just re-classified from "held-for-sale" to "held-for-investment."  HDFS took a charge to establish an allowance in Q2, which most sell-side analysts treated as entirely one-time charge, but I expect there will continue to be large provisions expenses every quarter. The sell side should figure this out when Q3 earnings are reported.  EPS estimates will come down, and the unexpected losses should stoke some real fear as analysts struggle to understand what is going on.  In addition, the fact that HDFS has had three presidents so far this year, (on top of the fact that HOG's CFO abruptly resigned in May without a replacement) signals that there could still be more problems to come for the unit.

I expect the Motor Co. earnings to recover to $1.75 in 2011 which assumes that the current free-fall in unit sales stops in 2010 and that the operating margin to recovers to historical levels in 2011 as a result of massive cost cutting.  I think it is worth a P/E of 12x to 14x resulting in a fair value from the Motor Co of $21-$25 per share in 1-2 years.  Arguably this is a generous valuation given the risks involved; it may take a lot longer for sales to stabilize, and HOG is cutting its operations by 25% which entails a huge amount of operating risk.  I assume HDFS is worth nothing.  This is very generous; if they were forced to mark-to-market its balance sheet today it would have huge negative value.     

 Description

HOG manufactures primarily heavyweight motorcycles 80% of which are sold in the US through a network of 680 independent dealers.  The majority (80%) of dealers sell HOG exclusively.  15% of revenue comes from the sale of parts and accessories and 5% comes from merchandise.  HOG has a cult brand with famously loyal customers and commands higher prices than competitors.  HOG dominates the heavyweight motorcycle segment with ~50% market share which is more than 3x the nearest competitor.  While HOG produces a variety of models, it is known for its classic styling and signature engine noise.  Its bikes have broad appeal, but within only one segment of the market, since it does not participate in light-weight or sportier segments.

HOG has a captive finance operation, Harley-Davidson Financial Services (HDFS), which finances approximately 50% of its bikes in the US (overall about 75% of purchases are financed), and it provides financing to its dealers.  Historically HDFS has contributed ~10% to overall earnings, but has accounted for roughly ~50% of HOG's assets and equity. 

HOG's average customer is a typical male baby-boomer, 47-years-old and married with median household income of $87k.  Importantly, the average age of HOG's customers has increased steadily from age 35 since the late 1980's, which will eventually provide a secular growth headwind.  Only 15% of sales are to customers younger than 35, who instead prefer sportier Japanese bikes which HOG has been unable to penetrate meaningfully with its Buell brand.

 Motor Business

HOG is one of the poster children of the consumer boom and bust.  In the 10-year period leading up to 2006 HOG's results where outstanding as it benefited from numerous tailwinds including: (i) increased motorcycle ownership industry-wide as motorcycles became adopted as a common "toy", (ii) increasingly wealthy and levered baby-boomer customers, (iii) increased market share, (iv) consistent prices increases, (v) mix shift into higher margin bikes, and (vi) rapid expansion of captive financing.

These tailwinds drove an 11.5% annual growth in units and 14% growth in revenue from 1997-2006.  Margins increased every year without exception, with the EBITDA margin increasing by an average or 130 bps each year.  HOG enjoyed tremendous operating leverage with unit sales increasing 170% over the period while the number of employees and manufacturing facilities increased by only 60%.  Most of these tailwinds have turned into headwinds: (i) unit sales dropped 13% from 2006 to 2008 and the company expects them to drop 30% this year, (ii) inventory is building at HOG and its dealers, driving price cutting with unit pricing down 5% year-over-year, (iii) accounts receivable have doubled as dealers take longer to pay, (iv) EBIT margins are down 400 bps over one year and 800 bps over two years, and (v) the percentage of bikes financed by HDFS has dropped 10% as HDFS seeks to stabilize its balance sheet.

The main drivers going forward will be the level of demand and the margins HOG can generate after cutting costs to adjust to a lower top-line.  The peak-to-trough drop demand is difficult to forecast given it is a big-ticket discretionary purchase that will largely depend on consumer sentiment.   However, the outlook is very poor given the new secular trend toward higher saving after a dramatic loss in household wealth.  HOG is having great difficulty forecasting the sales decline; at second quarter they dramatically lowered their 2009 guidance from a unit drop of -12% to -28-30%.  In my base-case I assume peak-to-trough demand drops ~50% from 2006 to 2010.  This is in-line with the 52% drop from 1979-1982, although that period was challenged by quality problems.  A 50% drop is also consistent with a reversion in sales per U.S. household back to late-1990's levels.  (Goldman Sachs has some data on this)

HOG is taking aggressive steps to cut costs including: (i) they are reducing 25% of the work-force in 2009 and 2010, (ii) they are cutting SG&A by 15% or ~$150 mm, and (iii) they are considering shutting or restructuring their York, PA facility, which is their largest.  HOG's cost-of-goods-sold is ~75% variable because it is mostly materials.  With the cost cuts and reductions in variable costs, HOG should be able to achieve a high-teens EBIT margin which is down significantly from peak EBIT margins of 24% in 2006.  But, it is similar to margins back in 1999 when HOG was producing a similar level of units. 

Financial Services (HDFS)

The finance business is poor.  Like the captive finance arms of the auto makers, HDFS exists to drive sales of bikes.  In the best of times it has a mediocre ROE, it relies exclusively on fickle capital market funding, it has significant credit risk, and the accounting is opaque and subjective.  The outlook for this business is very negative.  They have not been taking losses on $2.7bn of held-for-sale loans which are now going to stay on balance sheet, and for which they are going to have to start provisioning for losses like a bank.  The sell-side does not understand this and will be surprised.  Management does not inspire confidence.  Amazingly, HDFS has had three presidents since January.  This is on top of the fact that the CFO abruptly resigned in May, and the slot was only filled in the last few weeks by the interim CFO.

 Previously HDFS would sell its loans via securitizations and the loans would leave its balance sheet under "gain-on-sale" accounting treatment prescribed under FAS 140.  If the loans deteriorated, HDFS would write-down the value of its relatively small retained interest by discounting the cash flows associated with the retained interest.  The process for valuing the retained interest is highly subjective.  HDFS's recent TALF securitizations no longer qualify for gain-on-sale treatment (which is expected to be phased out in 2010 anyway).  As a result, in Q209 HDFS reclassified about $2.7 bn of loans from held-for-sale to held-for-investment. 

This classification change is very significant because the accounting rules for held-for-investment loans require HDFS to make ongoing provision expenses in order to create and maintain a balance sheet allowance for expected credit losses.  Under held-for-sale accounting, the loans were "Level III" assets and were valued using a highly discretionary DCF process.  HDFS clearly abused this discretion.  Amazingly, it only took at $44mm cumulative hit on $2.7 bn of loans since the credit crisis began.  The accounting change means it will be able to absorb the credit losses overtime via a provision expense, although it did have to take a modest $75mm charge in 2Q09 to establish a loan loss allowance for the loans it reclassified.

Credit costs will be significant at HDFS for at least the next two years due to high unemployment and declining residual values of motorcycles.  Credit losses have been running about 3% (annualized) in recent quarters on the entire pool of "managed" loans which includes all the loans that were previously held-for-sale.  Similarly S&P expects cumulative losses from recent securitizations to be about 8%, which is roughly equivalent to 3% per year given the short life of the pools.  The impact of now having to provision for loans that will be held for investment, means the quarterly provision expense will increase from $6-10mm per quarter to $40-$60mm per quarter, which drives significant losses for HDFS.  But contrast, most sell side analysts are assuming modest profits going forward.

Just for quick background on the liquidity situation, the securitization market shut for HDFS after Q108 so all the loans it originated began to accumulate in its held-for-sale account.  This balance ballooned from $700 mm in Q108 to $2.7 in Q209.  To finance this HDFS raised $1bn of medium term notes in 2H08 and $800 mm of commercial paper.   Then HDFS faced a near-death liquidity crisis in Q109 and it took several emergency steps such as raising $600 mm of debt from Buffett.  Subsequently HDFS has stabilized by accessing the securitization market through the Fed's TALF program, raising $1.5bn in two transactions in May and July, and another deal was just announced. 

Catalyst

Negative surprises from HDFS in the next few quarters, consensus estimate reductions, and retreat from current "full" valuation which assumes no risks of a recovery.