My BFBV course @ MDI got over in January 2014. One of the highlights of the course was a live case on Relaxo Footwear, a company in which I am invested. The case was initiated at the beginning of the course.
On 15 September 2013, I posted a mail (The Relaxo Cinderella Project) to my students about the company. At the time, the stock price of the company was Rs 144 (on a 5:1 split adjusted basis).
Then, on 22 September 2013, my friend Ravi Purohit and I gave a joint lecture on the company (The Relaxo Lecture) in which we explained our investment thesis. At the time, the stock was quoting at Rs 150. Finally, I spoke about the company again in my class on 10 January 2014. By that time, Relaxo’s stock price had increased to Rs 224.
As I write this, it now stands at Rs 254. In this note, I am reproducing from my memory what I spoke on my 10 January class with some updated thoughts on the subject.
You can find the transcript here.
22 thoughts on “The Final Relaxo Lecture”
Thanks a lot sir for posting another excellent post. Just few clarifications
1) The exit multiple of 15x, I think is based on reverse of post tax yield on 10% AAA rated bond. So in our exit multiple effectively we are not factoring in any growth, not even 2%.
2) Second I think the focus is on owner earnings and not PAT or adjusted PAT or free cash flow and by not deducting incremental working capital and maintenance capex, we are assuming that Relaxo will not need incremental investment in working capital and maintenance capex [not questioning the assumptions, just trying to understand the template]
Every template is created given the characteristics of the business. Working capital intensity has been reducing in Relaxo over the years, something I referred to in the earlier lecture. And maintenance capex is assumed to be a bit more than book depreciation. Both factors offset each other, so I kept it simple by focusing on PAT. In many other businesses that won’t work.
As for exit multiple, I used a “normal” multiple which implies moderate growth without thinking too much about how much that growth is going to be.
can we employ the same template to an NBFC which is enjoying strong entry barriers As far as exit multiple is concerned we can replace PE exit multiple with normal PB multiple. What I am not sure is what to take in owners earnings. I GUESS may be we can take PAT + provision for bad & doubtful debts minus credit costs over normal period [say last 10 years].
In a financial company, I would think about per-share book value and how that will grow over time. And of course, P/B ratio. You’ll also have to think about dilution as many financial companies issue equity frequently although that may not be applicable in the stock you may be working on. I’d be very careful about what P/B ratio to project 10 years out and would avoid going over 2 times and that too when I have assurance that prudence is built into the culture of that company and also that a premium over book value is warranted given a demonstrated track record of delivering high ROE while having excellent lending discipline.
Thanks a lot sir.
A small errata – in page 4, Rs 262 realized in 2023, and not 2013.
Thanks Prashant. Fixed.
The relaxo is selling 10 crore pairs in a year. Comparing with the population of India (120 crore) and economic condition of households, don’t you think that 10 crore is not a very low number. Moreover considering the company does not operate in leather shoes segment and limited presence in ladies footwears (sandals etc.) don’t you think it is close to maturing.
Obviously I don’t. If I did, then the expected return from current stock price would have been negative, no? 🙂
Thanks for the great lecture by emulating the framework and giving a easy to understand.
– It will be great to learn if this can be used for companies which are like railroads or tollgates (where the revenues will not be changing big and gross profit increase or optimization). Example — Not sure If I can try to use Kewel Kiran Clothing where there is no big change in revenue for the past 2+ years and this year as well.
– NBFC valuation — Got the answer above that use P/B < 2 times. May be good know if there is any classic reference to use for the future valuation. (example: Housing Finance)
You can use the method of focusing on expected return instead of fuzzy intrinsic value for any kind of business provided you have a good idea about what the business would look like a decade from now. That’s the first thing to keep in mind. The second thing to keep in mind is that you cannot use the same template for different businesses. The business model will determine the template. For instance in businesses which are capital intensive, your focus would primarily be on capital gains as there won’t normally be much of dividends in the returns equation. So the template could ignore dividends. In other cases, where businesses are not capital intensive, and dividend payout ratio is high, you must factor in dividend part of the return. In businesses where owner earnings and reported earnings are roughly equal, you may use reported earnings in your model. However, if owner earnings are materially different from reported earnings— which is often the case, then your template must allow for the adjustment. In some cases that adjustment would have to do with accounting depreciation vs. economic depreciation. In other cases, the adjustment may have to do with the changes in working capital intensity (reflecting power over suppliers and customers). In yet other cases, the adjustment may have do with spending on advertisement and publicity, and research and development.
Thanks for the wonderful article! It gave me a different, albeit logical, perspective of growth and earnings projection. A few questions:
1. Could you elaborate on the basis for choosing 15x earnings multiple?
2. How does one determine if a company can deliver the projected growth without selling additional equity shares?
3. What factors/ratios help one determine if the business passes the management test?
Sir..A humble request..You have always taught us about the art of buying. Can you please throw some light on Art of Selling.When and how to sell
At one point in his recent CNBC interview, Mr. Buffett talks about his mention of two real estate investments in his 2013 letter. He says:
“I came to a conclusion on that farm that it was likely to produce a little bit more over the years and that the crops would bring a little bit more over time and I bought it on a 10% yield-basis to start with.”
In other words, he focused on volume growth, realization growth, and starting cash flow yield of 10% a year.
It doesn’t get more fundamental than that.
[…] tried to answer those questions in a series of lectures on Relaxo Footwear which was a case in my MDI course which got over in […]
[…] Then, if that’s true, and I think it is true, you can reach out reduce the problem to just a handful of variables that will help you come with a range of potential future earnings and market values and from that you can derive a range of long-term expected returns. And I tried to explain that in my final Relaxo Lecture. […]
It has been really wonderful reading the article and should be very useful. Keep up the great work and One more request – whenever such lectures are there how can i participate – please inform.
sir, i had one question on periodic tracking using this framework. when you do analysis for a decade in say 2013 then with each passing year do you keep updating the end period value for the new 10 year period or do you keep calculating the returns with reference to original period end 2023. The reason i say this is that if you assume a certain volume growth for initial 10 year period, provided the penetration is still low, one could keep extending the growth to each new 10 year period. so in 2013 i might assume a high 10-12% volume growth for next 10 years with 15x exit multiple in 2023. Come 2015, do you assume similar growth for 2024 and 2025? Given low penetration, market growth and strong competitive position one could easily assume so.
Mr. Bakshi – While trying to get my head around owner earnings I have come to standstill at section titled “It’s The Owner Earnings That Count”. Off-course there may be lacunae in my understanding, but here are a few questions (all numbers below are in crores).
1. Advertising and Publicity expenses between 2003 – 09 have been excluded from the table. From annual reports annual figures (in chronological order) are: 8.9,12.32,13.16, 7.28,7.14,10.91,13.2
2. If these are included then Adjustment numbers start to look (from in and after 2005 – as I have limited the data to 2003 and as we are amortizing over 3 years we need to deduct charges each of past two years from add back of year in focus.): 1.7, (3.64), (2.05), 2.47, 2.78, 5.61, 5.32, 4.65, 18.56. [2005 to 2013]
3. If I consider 2014 A&P – 47.27, Adjustment comes out to 3.28. So the number of 2013 starts to look like an exception. And so would Adjusted PAT/pair CAGR not change?
4. Unadjusted PAT margin of 7.2 which comes to about 18.8 of PAT/pair is translating to Adjusted value of 21. From 6.3 to 21, implied CAGR seems to be 12.79, There is no clear relationship between PAT/Pair and Adjusted PAT/pair (if you include numbers of previous years, in some years Adjusted PAT/pair will dip below PAT/pair) how is one going to arrive at owner earnings number? Also the earning between 2003 to 2007 were jagged (but this may not be highly relevant here)
What is the function of translating 18.8 to 21(for this example and your recommendation in general)?
– Vishal Bhatia
Mr Bakshi – For valuation part, similar to most of your previous analysis you are using AAA bond benchmark rates of 10%, which is comparable to Pre Tax returns. Whereas valuation numbers of Relaxo – unlike your previous analysis like KKCL – have been arrived at by using PAT margins. Are you relying on pre tax yield benchmark to enhance MOS for valuation arrived by using PAT numbers?
[…] I am NOT suggesting that valuations are not important. But current or one year forward PE or PB multiple is not the correct method to decide about under or over valuation. One should follow the method suggested by Prof. Sanjay Bakshi in his Final Relaxo lecture. […]
Comments are closed.