Wednesday, 30 March 2016

Repost from March 4th - "H&R Block Down close to 20%"

I originally started writing this post back on March 4th, and about 1/2way through the post, realized that there was a serious deficiency in my analysis.  I'm reposting this as a walk through in order to compare my initial observations with what I think now is a bigger issue in performing valuations on the fly.  Hopefully this helps me bridge the gap between my incorrect initial observation and the correct set of conclusions to be drawn.

First, the original March 4th post (In Italics)

I'll do a more in depth post on HRB over the weekend, but here are my initial observations.

The stock is down close to 20% today on an earnings miss.  The culprit, as management explained it, was tax return filing deferral by taxpayers.  I guess this means permanent changes in the way taxpayers are going to file future returns equivalent to about $1.6B of market cap.

Here's a monthly chart going back to 1996.  Funny how in one session, HRB has wiped out close to three years of gains in terms of price.  The stock is back to early 2013 levels on today's move.

I've drawn two trend lines which I feel are important.  One catches most of the important lows from 1996 through current, with the exception of the 99/00 and 08/09, where HRB overshot along with the rest of the market.

The other catches the overshoot lows during the above corrective periods.

Here's what you get with HRB at close to a 20% discount to yesterday's closing price, and at almost a 30% discount from it's all time high made in late 2015, early 2016 (amazing how things can change so drastically in a span of a few months):

  • One of the most widely recognized franchises in tax preparation
  • Cash and securities of close to $2B ( = to approximately 30% of market cap).  Net cash is about $1.5B (almost equivalent to today's loss in market cap)
  • Pretty steady annuity type free cash flow generation (approximately $600M in free cash flow generated over the last three years on average)
  • A company with a small enough market cap at $6.5B (about $5B after net cash) to get swallowed up by an interested purchaser
  • A company in the midst of transformative change post disposition of HRB bank

By my rough calculations, $600M in free cash flow / .08 perpetually  = $7.5B, vs. $5B = about a 33% discount in a no growth FCF scenario.

And I promise, all you will read this weekend is how horrible a performer HRB was this week."

The problem with the original March 4th post in a nutshell

I was using 2015 reported #s to arrive at a conclusion in a hurry without properly analyzing what changed between April 30, 2015, and March 4th.  There have been some significant changes to the business since the last 10K, including:

  • Divestiture of H&R Block Bank (and along with it, bank capital restrictions) - see excerpts from investor conference presentation below
  • Addition of a significant amount of debt and upsized credit facility in order to facilitate a $1.5B tender offer to repurchase shares - see excerpts from investor conference presentation below

The Overall Problem With My Initial Observations

I mentioned in my March 5th post that by my rough calculations, perpetual free cash flow of $600M discounted at 8% would theoretically be worth $7.5B, and in comparison to the current EV (as at 04/30/15) of $5B, the valuation was compelling.

If this was the case, I'd be stupid not to buy the stock.  I'd be buying $.65 $1's.  But, as I've come to realize so very often over time, there is no free lunch in the market.

First issue

I didn't have my facts straight!  EV was not $5B.  It was closer to $9B post transition.  The company assumed a significant amount of debt in order to facilitate a massive share buy back north of $30.

Second issue

Once I realized the above, I did a quick back of the napkin valuation using the new #'s.  FCF of $600M / .08 = $7.5B.  Compare $7.5B to current EV of $9B, and all of a sudden, no more discount. In fact, based on my arbitrary evaluation rules, I was now looking at a 20% premium even after a -20% plunge.

This didn't make intuitive sense to me.  How could a valuation produce a -35% discount less than 1 year earlier when all that's really happened is a change in capital structure over the same period?

The answer, I think, was due to inconsistencies in my valuation approach.  And in order to solve my issues, I had to go back to valuation 101.  I've used a combination of Damodoran + CFA L2 slides to sort this out.  I've deferred to a free cash flow to the firm model, whereby the value of the firm = FCFF (FCF's + at interest x) / (WACC), assuming g =0.  I'm going to solve for PV FCFF under g = 0, and compare PV FCFF to EV.

My Hypothesis

No matter how I slice the capital structure, if a business hasn't fundamentally changed over a sample time period, there should really be no change in overall valuation (at least in the very short term). The fact that I came up with different valuations for HRB when the only aspect of the business that changed was capital structure post divestiture of H&R bank was a signal that I have been doing something wrong in my overall evaluation process.

Case in point, here are three sample valuation scenarios using different capital structures:

Case 1 - Mkt cap of $5B, excess cash of $1B, EV = $5B - $1B = $4B

FCF's of $500M and no interest expense  $500M / 9% = $5.556B value of the firm

Case 2 - Mkt cap of $4B, excess cash of $0B, EV = $4B - $0B = $4B

FCF's of $500M and no interest expense  $500M / 9% = $5.556 value of the firm

In this case, the company has used $1B of excess cash to buy back shares

Case 3 - Mkt cap of $3B, excess cash of $0B, debt of $1B.  EV = $3B + $1B = $4B

FCF's of $500M and interest expense of $1B x 2.5% x (1-.35) = $16M

$516M / X% = $5.556B?

In this case, the company has used $1B of excess cash + assumed another $1B of debt to buy back shares

If I use WACC of 9%, I get a higher overall value of $5.74B vs. $5.56B above.  In theory, this result is intuitively correct, adding debt to the capital structure up to an "optimal" level should increase the value of the firm, equivalent to at least the PV of the tax shield on the interest expense.

I need to brush up on my theory again.  For my simplistic purposes, I solved for X%?  = WACC of 9.29% in order for the value of the firm to be equal to $5.56B under all three cases.

This result also made intuitive sense to me, b/c adding significant debt to the capital structure should also increase the overall riskiness of the firm.

Somewhere between the two extremes, higher WACC or higher valuation, is an actual valuation (I just don't know what it is, and neither does anyone else).

Tying This All Back to HRB

Ok, so here's my revised work on HRB with the above in mind:

I looked at two cases on a FCFF basis:

  • Pre-transition, using an 8% no growth WACC model, I get a firm value of $8.2B.  At peak EV valuation back in the summer of 2015, EV was around $8.8B.  No discount.  
  • Post transition, solving for X% = WACC to get PV FCFF of $8.2B, I get 8.67%.  This compares to current EV of $8.4B.  No discount (even after a 20% drop)

Overall thoughts on the above results:

I'm leaning towards equalizing WACC's rather than solving for WACC's to equalize valuations, simply because I believe that adding debt up to an optimal amount should increase a valuation.  In this case, using 8% post transition, I get a firm value of $8.9B vs. current of $8.4B (and I still don't get a compelling valuation).

The problem, I think, with my initial approach, is co-mingling the concepts of FCFF and FCFE.

FCFE is much more difficult to pin down, because in theory, FCFE needs to be adjusted by the impact of net borrowings to get a value to capitalize at Re, and as a business goes through different life cycle stages, net borrowings change drastically.  In addition, capitalized FCFE should be the value to equity.  The capitalized result can't really be compared to EV, which is the value to equity + bondholders.

FCFF on the other hand, is a more appealing concept to me.  FCFF =  cash flows adjusted for after tax interest expense, and is more symptomatic of what's available to all stakeholders, equity + bondholders.  

The fact that I've been using FCF's as published by gurufocus without adjusting for after tax interest expense, capitalizing this result, and adding/deducting net debt probably means I'm farther off in my attempts at valuations than I would like to be  Back to the drawing board (again).

1 comment:

  1. Hopefully this helps me bridge the gap between my incorrect initial observation and the correct set of conclusions to be drawn.

    •°*”˜˜”*°•. H&R Block Recruits