Disclaimer: After the Cobrapost expose, there has been a lot of noise around whether the numbers reported by DHFL are genuine or not. While a rebuttal from the Management is expected, as a responsible student of valuation, I ask that you take this post with a mountain of salt. There's a famous quote in relation to Statistical Analysis that goes "Garbage in, garbage out". Put another way, if it actually turns out that the numbers were indeed misleading, then this post would be too. In that regard, I ask that you exercise due caution.
India recently witnessed something which could have possibly been its Lehman Brothers moment. The default of ILFS, a public sector infrastructure and transportation lending giant, caused a panic in the markets. NBFCs fell like a house of cards, and Dewan Housing Finance Limited lead the pack with a 40% drop in a single day. It fell even more in the following days. Did the market over-react? Let's take a step back and re-evaluate how much DHFL is really worth at this juncture.
What is 'Money and the Myth'?
If you have been a long-time subscriber to this blog, you might remember me making a post titled 'Numbers and Narratives: A Simple Discount Cash Flow (DCF) Model for Equity Valuation'. At the fag end of the post, I made a disclaimer saying that the model cannot be used to value BFSI firms and a Dividend Discount Model would be preferable instead. Well, this is it. 'Money and the Myth' is a simple Dividend Discount Model (DDM) aimed at valuing highly-leveraged BFSI firms.
Of course, like all my content, this one's free to download as well. The DropBox link to download the file has been provided towards the end of this post.
Happy Valuation!
The Company
Dewan Housing Finance Limited (DHFL) has almost three and a half decades of experience in Housing Finance in India, seconded only by Housing Development Finance Corporation (HDFC). Founded in 1984, DHFL caters to a niche segment in the Lower Middle Income (LMI) customers. With a gallery of awards reinforcing its strength as a company, DHFL is undoubtedly one of the premier Housing Finance giants in the country.
However, unlike HDFC, DHFL is prone to a minor suspicion. A few years back, rumors emerged that DHFL had cross-holdings in HDIL, a real estate promoter, whose founder was a blood relation of DHFL's MD and CEO. While the management has denied such claims of cross-holdings and other suspicious activities, we know for a fact what rumors do to a company's reputation.
Putting aside this dark spot for a moment, there's a similarity between HDFC and DHFL as well. Just like HDFC, DHFL holds a lot of subsidiaries--most of which are worth a lot. Here is the break-down:
 |
| (Source) |
The Industry
India's requirement for affordable housing has always been around. It was only in 2015 that it was reinforced by the government's launch of the Pradhan Mantri Awas Yojana. Of course, the ground reality is also quite similar. A survey done during 2017 showed that a majority of Indians prefer modest houses to expensive ones:
 |
| (Source) |
In line with the demand, Housing Finance companies have been seeing some amazing growth in the past few years:
 |
| (Source) |
Of course, one also has to consider the crisis caused by ILFS, which put pressure on the credit markets. Unlike a regular firm, BFSI firms are highly linked to the economic cycle. Rumors do affect the day-to-day business of these firms, punishing them with tight interest rates and decreased trust. Finally, the close relation to the economic cycle ensures that any small ripple in the economy will cause significant value creation and destruction in these firms. We will consider all these realities while trying to evaluate a company in this industry.
The Money
I have filled out DHFL's financial numbers as on Q2 of 2018-19:
Of course, as usual, the Net Income numbers are up-to-date, whereas the financial position numbers (Net Worth, Cash, Investments) are not, thanks to India Inc's poor reporting requirements. So treat these things with a pinch of salt.
If you go through the Balance Sheet of DHFL, you may notice that the company actually has a ton of cash holdings on its books (In Rs. Lakhs):
But this only underlines the need to look at the Notes to Financial Statements alongside the Financial Statements for any company. A bulk of this 'Cash and Bank Balances' are actually tied up in regulatory and margin requirements:
DHFL also has a ton of investments in short term bonds, thanks to the recent sell-off of its stake in DHFL Pramerica. As already shown, DHFL also has two subsidiaries. Since we are only considering the Standalone Net Income to value the company, we should add these figures to the final value as such (I have used a nominal P/B of 1.50 to value the subsidiaries):
You can take a look at how I arrived at the 'Company Beta' and 'Indexed Returns' in the file below:
 |
| (View/Download - DHFL Beta and Standard Deviation) |
Be informed that I have ignored the latest few weeks' data, which could be possibly distorted due to the credit crisis as discussed above. However, ignoring a few weeks' worth of data in 5 years' worth of data should not affect the outcome by a lot.
The Capital Conversions
DHFL has some Operating Leases and Equity Options on its book, which could possibly dampen its Value. The data to this extent has been taken from the 2017-18 Annual Report.
Operating Lease
DHFL has the following information regarding an Operating Lease on its books:
I have simply input these details into the model:
I have valued DHFL's Operating Lease at Rs. 5.50 Crores, which isn't a lot, I know. But in a valuation, one should leave no stone un-turned. For example, when I valued Five Below earlier, I found out that their lease-and-operate model meant that their appearance of a low-debt company was deceiving and they indeed owed a ton of money via Operating Lease obligations.
Equity Options
Talking about skeletons in the closet, DHFL has a close to 30,380 Options Outstanding at a price much lower than the current market value.
The good news is that these Options are supposed to expire this year and are the last tranche of outstanding Options on DHFL's books. I have valued them at Rs. 259 Crores as shown above.
The Assumptions
Unlike a DCF, a DDM requires a lesser number of inputs. Let me justify my inputs above:
1. High Growth Period: As discussed, DHFL is one of the oldest HFCs in India. With the credit crisis, RBI has been busy cancelling the licenses of several Housing Finance companies. This will only further increase the Competitive Advantage Period (CAP) of the existing, bigger players.
2. Convergence Period: This is the period during which all the inputs (Growth, Payout Ratio and Cost of Equity) will quickly converge towards their Terminal Period assumptions, in order to account for a smooth transition. The choice of numbers of years will hardly move the needle, but there's the option of choosing between 5, 7 or 9 years. For DHFL, I have chosen a 5-year Convergence Period.
3. Growth in Net Income: In the last decade, DHFL has managed to grow its Standalone Net Profits by almost 30% every year. But of course, since we're in the peak of the HFC growth cycle, that figure may be a little stretched. I could assume something like 25%, of course, but I would much rather go with the lower 19.24%, which is representative of DHFL's average mid-cycle growth from 2012-2016. Since I'm conservative with this estimate, I'm going to give some leeway in terms of reducing growth. I will reduce growth only by about 5% every year, as the growth figure slowly decreases towards the Self Sustainable Growth Rate and then towards the Terminal Growth Rate (Which is half of the long-term Risk-free Rate).
4. Payout Ratio: It is only logical that as DHFL becomes a bigger firm, it will find lesser and lesser opportunities to reinvest its profits. It will then pay out more and more Dividends instead. The Terminal Payout Ratio has been determined fundamentally (i.e. 1-Growth/Cost of Equity). This way, a higher Growth will be 'punished' by a lesser Payout Ratio, thereby maintaining a balance in the final Value of the company.
5. Discounting Rate: The Dividends will be discounted at the CAPM Company Beta. However, as DHFL matures, its Beta will move towards 1 (i.e. Market Beta) and so, its Cost of Equity will move towards the Indexed Returns (For further proof, look at the Cost of Equity history of some of the largest BFSI firms in the United States). I have simply emulated this economic reality.
The Dividends
Based on our assumptions above, this is how DHFL's Dividend Payments will evolve:
The last column also shows the composition of each Dividend in the final Value of DHFL. For instance, here we see that the High Growth Period Dividends form 47%, the Convergence Period Dividends form 21% and the Terminal Period Dividends form 32% of the company's value.
The Value
The moment of truth. This is how I value DHFL:
I believe that DHFL is fairly valued at about Rs. 433, signifying a 58% undervaluation. You will also notice that I have used a 30% Margin of Safety instead of the usual 10%. I did this because, as discussed earlier, the fortunes of BFSI firms are closely linked with the economy. With the entire industry in doldrums right now and the lack of clear direction, it is better to be excessively cautious.
In conclusion, I posit that the market indeed overreacted and the crash, to the extent it happened, was unwarranted. I also believe (Very similar to HDFC) that DHFL has a lot of potential to unlock value via its many subsidiaries. They did recently sell a portion of DHFL Pramerica Trust, which is why their liquid cash and investments balance is so high.
The Sensitivity of Value
The Sensitivity of Value tool shows the different Value ranges for the company for differing Terminal Growth and Terminal Cost of Equity values:
As I always claim, the Sensitivity of Value tool doesn't capture the entire range of Values in a stock. We have to resort to something more complex.
The Monte Carlo Simulation
I will now simulate DHFL's Growth, Payout and Cost of Equity estimates as follows:
1. Growth in Net Income: I'm going to simulate the Growth figures in the range of 13.13% (The Self Sustainable Growth Rate) to 25% (The historical average Growth Rate).
2. Payout Ratio: A nominal spread of +/- 15%
3. Cost of Equity: A nominal spread of +/- 15%
This is the initial result of a thousand (1,000) simulations:
After doing some Excel magic, this is how it can be presented better:
If you know Statistical Analysis and the Normal Curve, you will interpret this diagram like below:
After all, the true Value of a stock is not a single, fixed point. It is a range of Values for a range of future outcomes. At the current levels, therefore, it looks like DHFL has a greater than 90% probability of being undervalued. Is that enough? Only you should answer that question for yourself.
The Model
As promised, here is the link to download the model (For free!):
 |
| (View/Download - Money and the Myth) |
Of course, if your views differ from mine, feel free to download the model and re-value the company. Any significant divergences from my value can be discussed in the comments section below. I'm always up for a good, constructive dialogue.
When the Facts Change..
I have to admit here that I personally valued DHFL at about Rs. 580 not so long ago. So, what made me write down the value by 33% to Rs. 433? The facts changed. ILFS caused a near-extinction level event, the markets are spooked and the regulators are cracking down with extreme vigilance. The late, great economist Dr. J. M. Keynes once said:
When the facts change, I change my mind. What do you do, sir?
The mark of a good valuation is that it does not change a lot every year. It treads along the Cost of Equity/Capital imposed by the valuer. However, I have no shame in admitting that I should have been more cautious in valuing a company at the peak of a growth cycle. When reality hit, I changed my mind.
That was a good analysis
ReplyDeleteThank you for the kind words, Krishna Chaitanya.
Deletehi Dinesh
DeleteThanks for the hard work and analysis you have done and more importantly made is public without much of a self-interest !! May i request you to do a follow up on this article (maybe once a month) ?
Perfect. Not left with a doubt.
DeleteThank you, dhawal patel.
DeleteWow I have a company who has done a more detailed valuation, reached at a value of Rs.433/- apiece even after a 30% MOS, has 50% of it in his portfolio (i.e.,eating his own dog food), and doesn't regret about it. My back-of-the-envelope valuation was Rs.385/- apiece, and after 30% MOS, Rs.262/- apiece. Hence scooped at Rs.268/- for 30% of my porfolio on Friday. Will be happy to add more at a further 50% fall from here. Thanks Dinesh for the detailed analysis. Be greedy when others are fearful, and of course, change the mind when facts change. On a long-term perspective, the only glitch I had was the DHFL PRAmerica sale part which I feel is skewed in favor of the promoters even though the redemptions are 7 years away. What's your opinion on that?
ReplyDeleteHi Rajasekaran,
DeleteThank you for your comment.
1. DHFL isn't 50% of my PF. I increased my stake by 50%. As in, it was 10% and I raised it to 15%. I would mentally never let myself own more than 15%-20% of any BFSI firm, because of them being so highly leveraged.
2. I think the stake sale provided them with the much required liquidity they need. On the other hand, Insurance is an up and coming Industry in India. Perhaps it would have served the shareholders better if they held on to the stake. But I don't care much about non core activities as long as the core activity is being run effectively.
In the Assumptions-> Growth in net income, you mention " I will reduce growth only by about 5% every year". However, when I see your table, the growth in net income starts at 19.24% for yrs 1-5 and ends at 17% odd at 20 th year after which it falls to around 4% which is the terminal rate you have assumed. I did not understand how these assumed rates fit in with your statement of reducing it by 5% each year. Can you please explain?
ReplyDeleteHi Ganesh,
DeleteYes, I probably should have been clearer in expressing what I had done:
1. The Growth Rates I have used for every 5 years up to 15 years are 19.24%, 18.28% (19.24%*0.95), 17.36% (18.28%*0.95) respectively.
2. For the Terminal Year, Growth cannot exceed the Risk-free Rate. The Terminal Period represents time until infinity (Not actually, but at least in theory). So if a company grew more than the Risk-free Rate for several years (Say, 100), if would become larger than the economy itself, which is an absurd picture. So, it follows that a good practice is to limit Terminal Period Growth to the Risk-free Rate. Here, I have taken the Terminal Period Growth to be half of the Risk-free Rate, around 4.17%.
3. So, coming back to Growth in the 16-20 years, I have simply take a Simple Average of 17.36% (11-15 Years Growth) and 4.17% (Terminal Period Growth), just to smooth things out.
I hope this is better.
I got it now. Thank you.
DeleteHi Dinesh I have been holding dhfl since rs 23 . I haven't sold a single share. Right now it's a shorter s paradise. For me fair value was 385. I will anyway hold on. Unless of course it's something really drastic inside the company which we may never know.
ReplyDeleteThat is amazing! More power to you.
DeleteDinesh,
ReplyDelete2 points regarding your valuation.
1. Beta is a mathematical tool to estimate the riskiness of the underlying against the market and CAPM can be used loosely to value stocks. But if you listen to Prof. Damodaran's lecture on valuation on youtube, he essentially saying using CAPM for valuation is flawed since ERP itself is forward looking and cannot be estimated from past data. Hence, what I am trying to say is that investors will be demanding much higher returns than 18% a year to hold DHFL stock (since it is now considered very risky) so maybe the discount rate should be much higher. I am an investor in DHFL right now but I am demanding atleast 25% to hold the stock to compensate me for the risk
2. 20 year forward looking estimation is theoretically a good exercise but I use maximum of 10 years of forecast cash flows/dividend to estimate value even this with lot of trepidation. Don't forget current NCDs are trading at yields above 13% for DHFL - even if it falls to 11% DHFL NIM is 3% which means there is case to be made it cannot make any NIMs which means it loses it reason to exist. How can we be sure that it will be able to raise funds in the future at any kind of reasonable rate ie to generate an ROE above 15%
Hi,
DeleteThank you for writing.
1. I'm not sure about the context in which Prof. Damodaran said that, but he still uses the CAPM to estimate the Cost of Equity / Capital for all of his valuations. I'm not quite sure how you estimated that 'investors will be demanding much higher returns than 18% a year'. Beta indicates the market's demand for returns in a stock. But yes, like you (Who demands 25% to hold DHFL), anyone can demand anything to hold a particular stock. And the value to you or them will change accordingly.
2. Price is 100% in the past and Value is 100% in the future. So with due respect, I don't think a 10-year forecast is any different than a 20 or 25-year forecast as far as 'accuracy' is concerned. In fact, it is assured that we won't be accurate. That is why we demand a Margin of Safety. On the second part, yes, DHFL is bound to face short term pain. But the RoE used in the valuation is applicable throughout the valuation period. So I think it would be unwise to project the short term pain on to the long term story (You can try to manually adjust for the short term pain in the excel itself, but I highly doubt that will move the needle).
Hi Dinesh
ReplyDeleteI have been holding DHFL from past several years. My purchase is around 180 .After seeing 300% gain, It is now trading below the purchase prices :)
Lets assume whatever cobrapost has said is true. If there is any way individual investors can identify such systematic frauds, if it at all it turns out to be one
Unfortunately, even the best investors often make mistakes. So, we as regular investors will too, from time to time. The difference lies in whether we are willing to admit it and move on or not. I have added a Disclaimer at the beginning of this post to reflect just that.
Deleteyep. Now that I have waited for so long will wait until things get confirmed and take a call to stay or to exit.
DeleteBTW Dinesh.. Information which you have included here, is very detailed,helpful .Thanks for sharing it here on blog.
It's down a lot brealibr all assumption and prediction and calculation of valuation. I have been stuck at at higher level. Shall I sale or hold?
ReplyDelete