Thursday, 19 November 2015

Hibbett Sports Inc. and More Thoughts on My Valuation Model

I haven't had a chance to do a follow up post on Hibbett Sports Inc. (HIBB) as I've been travelling over the last two weeks.  It was my second recent buy in the bricks and mortar retail space, which seems to be unravelling daily.

I initially bought a very small position in HIBB, about $1,200 worth, at $33.50, for the following reasons:

  • I ran a two stage FCF DCF valuation and I thought the stock was cheap at the time
  • Around the time I bought the initial position, there was a news piece that Bass Pro Shops was mulling an offer for Cabella's.  I compared the two companies and thought that Hibbett had a much more interesting capital structure vs. Cabella's (due to much less leverage).  Cabella's rallied significantly on the Bass Pro rumour, and I thought that Hibbett was a logical acquisition candidate given it's market cap of less than $1B (pure speculation on my part)
  • Sentiment surrounding bricks and mortar retail was (and continues to be) abysmal
I was probably early in my purchase, which is fine.  Uncomfortable, but fine.  I can live with being wrong on $1,200 out of $130,000 in assets under management.

Before I get into further details, first, some comments on my DCF valuation and EPV modelling, as I continue to revisit and tweak the assumptions I've built into my in progress modelling.

Two Stage FCF DCF Matrix

I post a lot about running #'s through a Two Stage FCF DCF Matrix as part of my process. Essentially, what I've done is construct a lookup table which summarizes valuation results at different possible intersections of growth rates along the left hand side of the matrix, and required returns along the top of the matrix.  I do this because I have no idea what the future FCF growth is going to be (and I'm not smart enough to project).  What the DCF Matrix allows me to do though, is compare current market price to a range of possible valuations at different growth rates and different required returns.  If market price is lower than the worst case growth and highest required return, I may have a margin of safety.

Each cell in the matrix is equal to the intersection of a possible valuation result growing current year "normalized" free cash flows at growth rates over the first 5 years, next 5 years, and then terminally, and a required return.

For example, if I assume starting free cash flows of $59.05, yrs 1-5 G = 0%, yrs 6-10 G = 0%, and 0% terminal growth, discounted at an 8% required return, I get the following result:


G (0,0)0%0%0%0%0%0%0%0%0%0%0%
Year01234567891011
FCF59.0559.0559.0559.0559.0559.0559.0559.0559.0559.0559.0559.05
PV @ 8%54.6750.6246.8743.4040.1937.2134.4531.9029.5427.35341.88
Sum of PV's738.10
O/S Shares24.70
29.88
Net debt/sh-3.44
IV / share @ 8%33.32



Similarly, if I assume starting free cash flows of $59.05, yrs 1-5 G = 7%, yrs 6-10 G = 3.5%, and 3.5% terminal growth, discounted at an 8% required return, I get the following result:



G (7,3.5)7%7%7%7%7%3.50%3.50%3.50%3.50%3.50%3.50%
Year01234567891011
FCF59.0563.1867.6072.3477.4082.8285.7288.7291.8295.0498.36101.80
PV @ 8%58.5057.9657.4256.8956.3654.0251.7749.6147.5445.561,047.89
Sum of PV's1,583.52
O/S Shares24.70
64.11
Net debt/sh-3.44
IV / share @ 8%67.55



The culmination of the different lookups get summarized into a DCF matrix as follows:



Sum of DFC'sRe =
G = 8%9%10%12%15%Average, growth, blended Re
0,033.3230.0027.3523.3619.3826.68
1,034.6431.1528.3624.1820.0027.67
2,139.4634.9431.4226.3021.3730.70
3,1.543.2937.9433.8528.0022.4933.11
4,247.7841.3936.6029.9023.7235.88
5,2.553.1345.4139.7532.0125.0539.07
6,359.6050.1443.3834.3826.5242.80
7,3.567.5555.7647.6037.0528.1247.22
Average, Re, blended growth47.3540.8436.0429.4023.33
Current28.6528.6528.6528.6528.65



To me, the above DCF Matrix means a few things:

  1. At today's closing price, investors could be pricing in future growth in FCF's of between 3% and 4% over the next 5 years, and between 1.5% and 2% thereafter if investors require a 12% required return.  If FCF growth ends up being higher than 3% - 4% over the next 5 years, this growth should translate into a higher stock price
  2. Alternately, if HIBB continues to dissapoint or guides lower, investors could push the price down to reflect lower FCF growth expectations.  I'm pricing G (0,0,0 terminally) at $23.36 using a 12% required return, and $19.38 using a 15% required return.  Putting this in perspetive, at $20, the company would have a market cap of around $480M.  The company's cash  and inventory alone were $336M as at Aug 1, 2015.  Obviously this is an oversimplification though, because I'm not counting the premises leases and the liabilities. Total liabilities were $140M as at Aug 1, 2015, so cash + inventory less total liabilities  = $196M, or about 40% of market cap.
  3. Actual FCF growth over the last decade was about 10%, so maybe 3% - 4% growth is correct (maybe not).  If FCF was symptomatic of store growth, the company may very well have reached a saturation point.  No one knows what future FCF growth (or store growth) is going to be going forward, but in my layman's opinion, as the stock price moves closer to the intersection of zero growth and highest required return the less risky owning it becomes.
Obviously, the worst case scenario not being priced in is bankruptcy ala Radio Shack, but I'm guessing this scenario is unlikely, given that they sell specialty sports clothing/apparel, not 1985 toaster ovens.  Still, they must face intense competition from Amazon or Ebay (what else is new) and they have all the overhead that Amazon and Ebay don't in relation to premises lease commitments, staff wages, etc.

Another comment re: my FCF DCF Matrix.  I've started thinking along the lines of whether or not I should actually be recasting the entire P/L and cash flows in consideration of capitalizing the premises leases.  
   
For a comany like Hibbett, they had about $800M+ of premises lease commitments.  If I wanted to do a valuation taking these premise leases into account, I'd have to recast the P/L by adding back the rent expense, deducting a charge for interest expense (under the assumption that they bought and financed $800M of physical stores) and taking depreciation equal to the capitalized amount of the stores purchased/financed over their useful lives.  The end result would be probably be higher recasted net income as the imputed rent expense would translate into one part interest expense and one part principal repayment on the borrowings used to finance the stores. On a free cash flow basis, I'd probably have higher free cash flows due to a higher depreciation add back after the initial purchase of the stores.  However, I'd probably end up with a lower valuation because my net debt initially would need to be deducted up front.

Maybe what I need to do is model both scenarios and see what I come up with...stay tuned.  No one said this was easy.

EPV Analysis

I've also written previously about modelling earnings power value ala Bruce Greenwald's teachings. Basically, what I've done is use EBIT as reported and adjust for certain items which are expensed under GAAP, but which could theoretically result in future cash flows.  Greenwald gives a couple of examples in his teachings concerning WD40 and Intel, and suggests that a portion of R&D and SG&A should be added back to EBIT.  He used 25%.

Here's what's bothering me though:  25% seems arbitrary.  While adding back 25% of R&D and SG&A probably makes sense in relation to Intel or WD40, why would anyone think of adding back 25% of SG&A for Hibbett?  They're a sports footwear / equipment / apparel retailer!  Included in SG&A are things like: store operating, selling and admin expense, and advertising.  Included in cost of sales are things like: logistics and occupancy.  I'm not sure how any of these line items would induce future sales!  They are period costs.  Maybe, an argument can be made to add back 100% of the company's advertising expense, but total advertising was only about $6.3M for the year ended Jan 31, 2015.  A far cry from 25% x total SG&A of $200M.

In any case, I'm going to present the results of the EPV analysis using a 25% SG&A add back and a $6M SG&A add back for comparative purposes:

First, 25%  x SG&A



Cost of capital rates:EPVAdj for Debt & Excess CashAdj EPVO/S sharesEPV / shareBVPS / shareExcess EPVCurrent pricePrice / EPVPrem/Discount vs EPV
Stress test15%675.5585.00760.5524.730.7913.272.3228.650.93-6.95%
LT equity return; 2014 - 192811.53%878.8685.00963.8624.739.0213.272.9428.650.73-26.58%
Upper bound10%1,013.3385.001,098.3324.744.4713.273.3528.650.64-35.57%
Combined NI & Div growth9.24%1,096.9085.001,181.9024.747.8513.273.6128.650.60-40.13%
CAPM7.00%1,447.6185.001,532.6124.762.0513.274.6828.650.46-53.83%
Per GF4.97%2,038.8885.002,123.8824.785.9913.276.4828.650.33-66.68%



Second, $6M SG&A
 


Cost of capital rates:EPVAdj for Debt & Excess CashAdj EPVO/S sharesEPV / shareBVPS / shareExcess EPVCurrent pricePrice / EPVPrem/Discount vs EPV
Stress test15%491.8085.00576.8024.723.3513.271.7628.651.2322.69%
LT equity return; 2014 - 192811.53%639.8185.00724.8124.729.3413.272.2128.650.98-2.37%
Upper bound10%737.7085.00822.7024.733.3113.272.5128.650.86-13.98%
CAPM7.00%1,053.8685.001,138.8624.746.1113.273.4728.650.62-37.86%
Combined NI & Div growth5.82%1,266.4985.001,351.4924.754.7213.274.1228.650.52-47.64%
Per GF4.97%1,484.3185.001,569.3124.763.5313.274.7928.650.45-54.91%



Completely different analysis!  In the second (more conservative case), Hibbett only appears to be about 14% undervalued relative to EPV using a 10% WACC, a far cry from the 35% indication under the 25% SG&A case.

Finally, a comparison to other specialty retail companies:



















Concluding Thoughts

I'm sort of scratching my head here as the company seems cheap across a number of metrics, and I'm beginning to wonder if I've missed or overlooked something significant in my analysis.  I guess time will tell...


 

No comments:

Post a Comment