ValuePickr Forum

Expectations Value

After using most of the standard valuation techniques and back testing them on stocks, I realized that none of them work the way we expect. The predictions are always wrong and the response from the stock market is well, always contrary to what we expect.

So I reversed the thought process and started trying to understand, why does stock market reward certain stocks in certain years and not do so for others which seemingly have same performance. I eventually converted this thought process in my own valuation model, “Expectations Value”.

Market in its collective wisdom values a company on the expectations of its earnings growth for the current year. Meeting or exceeding these expectations result in the markets rewarding these securities. This is similar to employee appraisal process.

I have extensively back tested this model and the results are also available on my blog. Inviting comments from this esteemed group. Thank you.

6 Likes

That’s a very innovative way to look at valuing a stock. I’ll give you that.

But let’s address the elephant in the room. Hindsight Bias is big in the investing world and I should assume that most would agree with me. You can look back at any stock in any industry and build a very convincing story about how things would have played out eventually.

But we do not have that luxury in investing. In fact, that is the whole point of investing in the first place - risk foregoing your investment for a shot at earning more money on it. Clearly, you have just placed your thesis here, and I do not want to appear dismissive. But if possible can you answer the following two questions:

  1. How do you determine the expected Earnings of a stock, say, 5-7 years into the future?
  2. Once that is done, how do you determine whether or not the company behind the stock will underperform or overperform the expectation?

Clearly, there are even more subliminal questions here, such as:

  1. What about cash flows? Certainly, a company can earn “profits” for 10 years and still end up producing 0 in terms of actual cash.
  2. What is the logic behind the sample set chosen for the proof of concept? Did you consider only specific stocks or did you consider the entire listed universe? Why?

You can answer these too if you like, but I’d say the first two are the most fundamental.

Thank you Dinesh for your response. I will try and answer your queries in 2 posts.

The model claims that if a stock outperforms its expectations, it is better placed for greater returns (as compared to market returns). My blog details out the proof of concept and case study to make this point.

Per the current structure, stocks are shortlisted post results and are held for an year, just like the magic formula. Once the results are out for the next year, a different set of stocks is shortlisted for investment. Hence the expected earnings need to be determined only for the current year. So in May 2019, the model calculated the expected earnings for FY2019-20. I can provide the expected earnings growth for Nifty 50 stocks here.

I cannot disclose the entire process of calculating the expected earnings, but it is based on fundamental analysis and statistical methods.

Once the expected earnings are known, the critical aspect is viability check. Again this is based on fundamental analysis and company’s historic performance. The blog proof of concept only shows a list of companies which have met the expected performance (using the benefit of foresight). In real life, it is impossible to achieve this kind of performance. It is just a PoC for the idea.

My real life back test (with no use of any future data) produces over 25% CAGR for large cap (top 200) portfolio for the last 14 years of testing. My May-2018 large cap portfolio has also produced over 20% returns as on date.

The model is based on fundamental analysis, but it focuses more on the earnings as compared to cash flows. The cash flow based filters are applied post the model shortlist during the viability check.

The proof of concept was based on the entire stock universe. Over 1500 stocks fit the model, rest are filtered out. The “actual Vs expected” check was applied on these 1500 stocks. Of course, the PoC had the benefit of foresight and the returns are just a proof that the model works.

Well, yes. And why do you think that is the case? If the Proof of Concept clearly shows that it has been massively successful in the last 16 years, why not for the next 16 as well?

In short, will you bet your entire wealth (Or whatever wealth is in equities alone) on religiously following this model?

You missed my point on the real life test.

Yes, I am investing purely based on the model. I will disclose my 2019-20 portfolio in the first week of June.

2 Likes

This sounds quite interesting. Your buys are once the FY results are declared? Do you also rebalance based on quaterly results?

Quarterly re-balancing is not required, unless an exception case like fraud is disclosed. Almost the entire portfolio is bought (and sold the next year) in April-May when the yearly results are declared. There are a few cases when the companies follow a different FY calendar and the model also gives the recommendations accordingly.

Sure then… why not. I have always been intrigued by Magic Formula-like automated investing models. I guess we will see how this turns out, say in 3-5 years.

1 Like

Nice write-up on the blog. Just wanted to know if you can share how you are deriving on the expected earnings it will be good.

Also, you are keeping a horizon of just one year. If the company is in growth stage it may give good returns in the subsequent years as well. In that case why one will make a decision to sell that company. Can you throw some light on this aspects?

Apologies, I cannot share the methodology I use to arrive at the expected earnings growth. But I can share the expected earnings growth of any company you need.

If the company is in growth stage, it may be shortlisted in the next year. If it does, we will not sell it. But in our back tests, such occurrences are rare. Typically, when a company grows faster than expectations (like HEG, Graphite, Shivalik), the stock prices rises so far ahead that the next year expectations soar beyond reality. Holding these stocks for the following year usually results in loss. It is hence important to hold the stocks at the right time.

Can you please share expected growth rate of below companies

  1. HDFC Bank
  2. Yes Bank
  3. Avanti Feeds
  4. Aavas Financiers
  5. Laurus labs
  6. Syngene International
1 Like

Sorry to intervene.

I think it’s better to DM the OP for details like this. I’m afraid if OP starts posting expected growth rates here, we’ll get a cluttered thread full of expected growth requests and answers.

Let’s make sure we don’t deviate from the purpose of this thread. Use DMs for data requests.

1 Like

The objective of asking the expected earnings growth is not for any guidance. I just wanted to check the thesis and was prepared to ask questions based on the reply. As we all know the HDFC Bank is growing at around 20% so the answer for the first is simple but for other companies it becomes difficult and if any thesis or model solves that problem then it should be considered for further analysis.

1 Like

I will post the expected growth rates for all NIFTY 50 stocks in the first week of June. For the ones listed above, Aavas, Syngene and Laurus do not fit the model (mostly because of limited data availability, recent listing). Avanti is yet to publish the result for 2019.

Expected growth for the rest:

HDFC Bank - 20.37%
YES Bank - 19.25%

I agree. Though I have responded to Nityanand, I will publish the Nifty 50 stock expectations on google spreadsheet and give the link here. If I get more requests thru DM or on this thread, I will respond by adding those stocks in the spreadsheet.

1 Like

Just a quick question, if the Yes bank comes up with more NPAs in next few quarters then the growth rate which you are mentioning will not be possible then how your model calculates the future growth in that case?

All I can say is that the provisions do impact the calculation of future expectations for banking and finance companies. Expectations are derived based on the price and performance.

I have published two new posts:

  1. Expected Earnings Growth for the NIFTY 50 stocks
  2. EVM Portfolio for 2019.

I have also updated the 03. Case Study - Nifty 50 - 2012 - 2018 post to include historical data from 2012. This will give a very clear picture of how the EVM works.

Do go thru the posts and let me know your feedback.

1 Like

Portfolio performance for the 4 months (June-Sept 2019) updated on the blog:

In a nut shell: Model portfolio of 2019 is down about 40 basis points. You can check the detailed performance here. This excludes the dividends we have earned and also misses out on Aarti Surfacants (de-merged from Aarti Industries), which is yet to be listed. Post these adjustments we will probably be just in the positive.

1 Like