Ideal Capital Structure For the Auto Retailers

by: Jerry Marks

The Modigliani-Miller theorem (of Franco Modigliani, Merton Miller) forms the basis for modern thinking on capital structure. The basic theorem states that, in the absence of taxes, bankruptcy costs, and asymmetric information, and in an efficient market, the value of a firm is unaffected by how that firm is financed. It does not matter if the firm's capital is raised by issuing stock or selling debt. It does not matter what the firm's dividend policy is. Therefore, the Modigliani-Miller theorem is also often called the capital structure irrelevance principle.

Merton Miller's analogy to illustrate the principle uses a pizza: cutting a pizza into more or less pieces does not change the underlying amount of pizza.

Source: Wikipedia

Capital Structure part 2

Last week I asked the question "what makes the ideal capital structure?"

I gave you some of the comments I heard from leading CFOs and Treasurers in the space.  And then I kind of left you hanging when I said "I think Brian (Campbell, AutoZone's treasurer) and Jeff (DeBoer, Lithia's CFO), are really onto something when they talk about it depending on the cash flows the company will receive. 

Harry Yanowitz (Pep Boys CFO) also made comments on Pep Boys earnings conference call last week suggesting the company's capital position depended on its cash flows when asked about sale-leasebacks and capital structure.

So today I wanted to show you what the operating "coverage" of each of the companies in the index looks like. 

Before I do, however, I just wanted to explain that there are a number of ways to look at interest expense coverage.  But most finance folks look at two metrics: EBIT, and EBITDA. 

EBIT is simply earnings before interest and taxes (operating income).  So your sales minus whatever it cost for the product (including labor if it is a service you are selling), and subtracting out any selling, general, and administrative expenses.

You can then arrive at an EBIT coverage ratio by simply dividing the interest expense.  For example, suppose you generated $100,000 in EBIT, and had $20,000 in interest expense, you have an EBIT coverage of 5x.   Meaning, in theory, your EBIT would need to fall by more than 80% before the operating income being generated by the company would no longer be able to fund your interest expense payments. 

On the other hand, if you had an EBIT of $100,000, and interest expense of $95,000, which works out to a coverage ratio of 1.1x, a slight drop in your operating income could make it difficult for you to fund your interest expense (at least from internally generated operating income). 

EBITDA expands this concept.  Because operating income may not give finance people the real skinny about a company's ability to fund its interest expense.  EBITDA is technically the earnings before interest, taxes, and depreciation and amortization. 

When you buy a computer (building, or any other piece of property or equipment), the accounting rules say you need to depreciate it over a set period of time (and this gets recorded in the income statement either under cost of goods sold or general administrative expenses). 

Say you spent $100,000 on a bunch of computers for the office.  The accounting rules might say you need to depreciate the equipment over three years.  So every year, you would be recording $33,000 in general and administrative expenses. 

But the money was spent upfront.  So you really aren't spending "depreciation and amortization dollars."  It is simply an accounting expense. 

Therefore to try to get a better picture of the cash a company is generating to cover the interest expense, finance professionals add back the depreciation and amortization (hence EBITDA).   

Now there are also changes in working capital (inventory, payables, receivables), that complicate things more (and why some people do not consider EBITDA a good measurement of cash flows). 

For the purpose of just trying to get a temperature of what each company's capital structure looks like (relative to its earnings/cash flows), however, let's not get too technical and stick with EBIT and EBITDA.       

So the table below illustrates each company's trailing twelve month EBIT, EBITDA, and interest expense and then the "coverage" ratio.     


($ in thousands)

EBIT EBITDA Interest expense EBIT coverage EBITDA coverage
Lithia $111,401 $130,484 $18,251 6.1x 7.1x
Group 1 $198,540 $217,808 $22,833 8.7x 9.5x
Asbury $167,922 $188,759 $39,702 4.2x 4.8x
Sonic $287,920 $314,331 $36,647 7.9x 8.6x
Pep Boys $50,148 $88,746 $52,024 1.0x 1.7x
Penske Automotive $327,258 $376,992 $58,200 5.6x 6.5x
Genuine Parts $869,705 $949,893 $24,702 35.2x 38.5x
Keystone Auto Parts $58,162 $68,558 -114 N/A N/A
O'Reilly Auto Parts*** $158,750 $228,889 $3,841 41.3x 59.6x
Midas* $30,000 $39,000 $9,000 3.3x 4.4x
LKQ Corp. $90,127 $103,102 $7,360 12.2x 14.0x
Monro $42,600 $63,151 $4,968 8.6x 12.7x
America's Car-Mart $10,406 $13,400 $3,728 2.8x 3.1x
CarMax $341,311 $378,622 $7,389 46.2x 51.2x
AutoNation $754,500 $837,400 $53,753 14.0x 15.6x
AutoZone $987,977 $1,121,505 $105,778 9.3x 10.6x
Advance Auto Parts $419,109 $567,108 $35,743 11.7x 15.9x
CSK Auto Parts** $63,910 $110,658 $58,676 1.1x 1.9x
Hertz $1,155,899 $3,033,481 $860,049 1.3x 3.5x
Copart $196,482 231,778 72 2,723x 3,219x
Click to enlarge

Source: company reports, efficient insights llc

***O'Reilly's interest expense is a best estimate based on annual interest expense reported in 2006 and "other income/expense" reported in the quarterly reports.

**CSK operating profit excludes investigation costs

*Midas operating profit excludes business transformation charges

A couple comments about the above table: I tried to adjust out extraordinary items that occurred in the operating results over the last 12-months where possible.  But mistakes happen and I may have missed some.  Management teams are welcome to point out any mistakes and I will correct. 

Also, in most cases, I used the interest expense line item (not netted out for interest income).  I did this because some companies report interest income in "other income" (meaning other profits and losses from things like gains on investments are also in the figure).  Unfortunately, I could not get an interest expense figure from Keystone's filings and therefore was forced to use the "other" (income/expense) line item. 

Finally, when it comes to franchised car dealers, you need to keep in mind that floor plan borrowing kind of works as both financing (capital structure) and a working capital/operational issue.  For those of you not familiar with the dealer world.  Floor plan borrowing is simply what a dealer borrows (usually from an automaker) to fund their inventory.   

Because floor plan borrowing is tied to inventories, those of us analyzing the dealers really should consider floor plan borrowings a "trade payable" kind of similar to any other working capital item. 

However, we also must recognize that sometimes management teams "park cash" in the floor plan line. 

It works like this: CFO of company XYZ decides he/she will need $50 million for an acquisition or to pay down a long term note coming due in the next 6 months (for example). 

But if the CFO just put the money in the bank (or in short term treasuries) they might only get 3% or 4%.  Whereas paying down the floor plan debt (even though it is a pretty cheap form of financing) temporarily allows them to reduce floor plan interest expense that they are paying 5%/6% on (so a better use of the cash in the short run).   

So in order to give you an idea of the floor plan "utilization" rates, below are the franchised auto retailer's floor plan borrowings as a percentage of total inventories (including the parts sitting in the service bays). 

I also threw in things like new vehicle inventory as a percent of total inventory, and floor plan "trade" as a percentage of total floor plan debt. 

The floor plan "trade" simply refers to floor plan debt financed through an automotive manufacturer versus some independent finance organization. 

I think the public dealers have struggled with how to count GMAC, which is now 51% owned by Cerberus.  AutoNation counts GMAC as "non trade," but I think some of the other large dealer groups still count GMAC as trade floor plan.  So you should keep this in mind when looking at the figures.  

Metric Asbury AutoNation Lithia Group 1 Penske Sonic*
Floor plan as % of total inventories 82.6% 89.5% 83.7% 84.7% 95.5% 102.2%
Floor plan borrowings per vehicle (retail) $15,313 $15,131 $19,307 $13,770 $19,940 $21,133
Floor plan borrowings per vehicle (retail and wholesale) N/A N/A $15,785 $11,148 N/A $17,321
Floor plan trade as a % of total floor plan debt 29.7% 78.3% 70.3% 18.5% 71.0% 22.7%
New vehicle inventories as % of total inventories 78.5% 78.0% 78.4% N/A 71.6% 79.4%
Click to enlarge

Source: company reports, efficient insights llc

*Sonic's floor plan as a % of total inventories would be 94% if I include assets held for sale.

Earnings yield

So now that you have a little better picture of the risk being taken by each company to generate that "cash on cash" return on investment.  Below are the earnings yields for the companies in the index. 

Remember, earnings yield is simply the earnings per share (after all expenses and taxes), divided by the share price (what you would have to pay for the stock based on Friday's close). 

Company Current Year Estimates Next Year Estimates
Lithia 9.7% 11.7%
Group 1 10.7% 11.7%
Asbury 10.4% 11.5%
Sonic 9.5% 10.5%
Pep Boys 2.3% 3.6%
Penske Automotive 7.3% 8.2%
Genuine Parts 6.2% 6.8%
Keystone 4.5% 5.3%
O'Reilly Auto 5.0% 5.8%
Midas 4.4% 4.9%
LKQ Corp. 3.2% 4.0%
Monro 5.0% 5.9%
America's Car-Mart 6.0% 8.9%
CarMax 4.5% 5.4%
AutoNation 7.7% 8.6%
AutoZone 7.0% 7.7%
Advance Auto Parts 6.6% 7.6%
CSK Auto 6.1% N/A
Hertz 4.5% 6.5%
Copart 4.9% 5.4%
Click to enlarge

Source: Zack's, Yahoo Finance, efficient insights llc

Now last week when I showed this earnings yield table, a good friend of mine emailed me (I won't hold it against him that he is at a hedge fund), and said that earnings don't give a good picture of true cash flows because some companies (for example) have depreciation expenses from businesses that are no longer a part of their strategy (i.e. Midas depreciating its exhaust business). 

I agree.  But I have also discussed how changes to working capital (particularly with car dealers floor plan borrowings) can leave looking at free cash flows more misleading than looking at the reported earnings.  This is why I tend to focus your attention on the earnings yields

Capital expenditures as % of depreciation and amortization

Having said that, looking at earnings yields requires a basic assumption: that the depreciation equals "maintenance" capital expenditures (purchases of property and equipment).  And "cap x" beyond depreciation should therefore be investment for growth. 

So when it comes to most "mature companies," I usually like to see capital expenditures below reported depreciation levels (because I also think that new capital expenditures should be invested more efficiently/effectively). 

"It's not the dollars you spend, but how you spend them" (I always try to tell people). 

Therefore in order to help you understand the true earnings/cash flow picture, below are the trailing t),welve month capital expenditures, depreciation and amortization, and cap-x as a percentage of depreciation and amortization. 

Company Cap-X Depreciation and Amortization Cap-X as a % of Depreciation and Amortization
Lithia $92,167 $19,083 483%
Group 1 $97,307 $19,268 505%
Asbury $50,839 $20,837 244%
Sonic $84,646 $26,411 321%
Pep Boys $52,434 $38,598 136%
Penske $187,341 $49,734 377%
Genuine Parts $120,219 $80,188 150%
Keystone $13,658 $10,396 131%
O'Reilly $21,186 $70,139 30%
Midas $3,500 $9,200 38%
LKQ Corp $37,831 $12,975 292%
Monro $21,191 $20,551 103%
America's Car-Mart $2,716 $994 273%
CarMax $227,504 $37,311 610%
AutoNation $200,500 $82,900 242%
AutoZone $239,172 $133,528 179%
Advance Auto $242,223 $147,999 164%
CSK $38,457 $46,748 82%
Hertz $335,575 $1,877,582 18%
Copart $96,673 $35,296 274%
Click to enlarge

Source: company reports, efficient insights llc

I'll let you interpret the data as you may.  But I just wanted to say one thing.  The franchised auto retailer's have cap-x as a % of depreciation and amortization levels greater than some of the fastest growing companies in the sector (like LKQ). 

This suggests to me that the cap-x dollars they are investing (often times required by the manufacturers) is not generating a sufficient return. 

I can only encourage the public dealer management teams to work very hard with their manufacturer partners at helping them see more efficient uses of the capital (that will generate higher sales and market share) versus being wasted on facility upgrades/improvements that are clearly not achieving adequate returns.