Ever since I introduced the concept of SI-PE (PE based SIP investing) in the post, '**Are Mutual Fund SIPs Suitable for Disciplined Long-Term Investors?', **I have** **been worried because it instantly struck a chord with many readers.

SI-PE refers to manual investment based on the PE value of an established index. If the PE is greater than 20 (value chosen arbitrarily! ), monthly investments are put on hold. All the uninvested amount is invested in one go when PE drops below 20.

In the same post, it was shown that a SIP (blind monthly investing regardless of the state of the market) works perfectly for the long term investor. SI-PE scores over the SIP but only by a little.

Since then I have received many questions on the SI-PE, leaving me worried.

Is the Nifty PE really relevant for the long-term investor? Of course, if I invest when the PE is 28 and look at my holding a year later, I am almost certain to have lost money. What if I chose not to touch my holding? What if I held on to it for say, 10 years?

Would it still make a difference?

In this post, let us try and answer this question. I present below a series of plots of the Return obtained versus Nifty PE corresponding to the investment date for durations ranging from 1Y to 13y. The data was obtained with the **rolling returns calculator**. Nifty closing values and PE from 1st Jan 1999 to 30th June 2014 were used for the analysis

If you need the Excel sheet, leave a comment.

These plots are wonderful examples of market volatility. So I suggest you stare at them for long periods of time!

**1-year**

The vertical blue line represents 0% returns! The horizontal red line represents the range of returns possible for a give PE value. For example, the current Nifty PE is a little less than 21. So if you invest a sum now, the return after a year could range from -50% to +60%. Just about anything.

However, if the PE was 25 or more, The return is likely to be negative (-50% to -10%).

Interestingly if the PE was close to 10, the return can still vary by a huge margin (+30% to +90%)

Notice that the bunch of points slope down from the top left to the bottom right of the graph. This is referred to as a *negative correlation*: Higher the PE, lower the return. The *width* of the bunch represents the extent of correlation. That is, if the range of returns possible at a given PE is narrow, the (negative)correlation is strong.

## 2-year

Notice the range or returns possible for current PE levels (red line). Again just about anything. Higher the PE, lower the return, but the range of returns has increased.

## 3-year

Notice that the bunching has disintegrated. The points have spread out and the negative correlation has reduced. Returns for a sum invested 3 years ago at a PE ~ 11 can just about be anything, 5% to 60%! Investing at high PE levels still implies a negative return.

## 4-year

The negative correlation has lowered further. The range of returns possible for investing at low PE has increased 🙁

Notice that the bunch of points is moving toward the right - more positive returns.

## 5-year

The bunch has now moved almost entirely in the positive region. Surprisingly the negative correlation has strengthened now. That is the bunching is tighter now. If one invests at a PE level of 25 and above, it would take about 5 years to get a small but positive return!

So the case for *not* investing at high PE levels is very much alive!

Surprisingly the case for investing at low PE levels comes tagged with a huge spread in the possible 'high' returns.

## 6-year

The plot looks like a swordfish! Again the bunching has reduced for low and medium PE values. The bunching is tight at high PE levels (the sword!)

## 8-year

The bunching has reduced again! Investing at PE levels of 15 and lower has very high chance of getting a double-digit return. Not enough high PE level data points. Whatever exists completely contradicts the high-PE equals low-return theory!!

## 10-year

The bunching and negative correlation returns!

## 12-year

The correlation is stronger, but the number of data points have reduced significantly. Those who invested at PE levels of 25 and above, 12 years back would be sitting with a return of about 10%, while those who invested with Nifty PE less than 15 would have got about 15% or more.

Whether this is significant or not can be debated endlessly. As mentioned above, the case for *not *investing at high PE (~ 25) remains strong enough, whereas the case for investing *only* at low PE (< 15) has weakened, thanks to the wide range of returns possible.

Should one forget about all this and just do a mutual fund SIP? Not a bad idea at all.

Is SI-PE worth it? well, yes and no.

Yes it makes sense not to invest when the PE is high. Yes it even makes sense to pull out a portion of the equity corpus when the PE is very high (> 25).

However, it also makes sense to get rid of the salary into an investment each month. So perhaps one can have a base SIP running and 'time' the additional amount according to the PE. Perhaps.

It all comes down to ones mental framework. If one can remain cool, calm and stop worry about the grass on the other side of the meadow, SI-PE will work.

If one gets worked up each time the PE increases by a few points, SI-PE is not worth it.

I do monthly manual investing and do try to invest on 1-2% market dips each month.

On the one hand, my monthly investment is high and my portfolio is getting fatter, meaning soon it will swing by 60K-100K over a few days if not over a day. So I cannot simply invest more or let the money lie around when the PE is high. I however have a lot of inertia and will not do this impulsively.

On the other hand, I am an investment junkie. Unless I put away my salary as soon as possible, I will feel restless. The money may or may not get spent, but I will quite uneasy. So I will need to balance this.

What I do may or may not be smart. It is *my *thing. You will have to figure out yours. If you are a 'new' investor stick to a SIP!

Shyam Sunder BonguNice Article. Thank you. One basic question. How could we know the P/E of Nifty?

Ramakrishnahttp://nseindia.com/products/content/equities/indices/historical_pepb.htm

pattuPost authorThank you.

CraytheonYou can see the long term Nifty PE chart here

http://craytheon.com/charts/nifty_pe_ratio_pb_value_dividend_yield_chart.php

pattuPost authorHi guys welcome!

craytheonHi Pattu, I have been browsing your website for a while, you have created a fantastic website specially the various calculators,

pattuPost authorThank you.

Nihar PalSuperb analysis, explained well. Thanks

pattuPost authorThank you Nihar.

ShanI think this kind of analysis needs to be refined further. There are 2 components to P/E, the "P" (price) and the "E" (earnings). If you assume that earnings remain constant, then buying when P/E is low and not buying when P/E is high makes perfect sense. But the analysis you give breaks down when "E" changes! When the earnings fall unexpectedly, you will have a very high P/E and you will in fact sell off your stock. But earnings tend to rebound quickly so you would have sold off at exactly the wrong time and missed on the upside. What applies to a fall in earnings also applies to a rise. High P/E is justified if earnings rise sharply.

Bottomline: P/E is a backward looking indicator and you buy stocks for the future. Don't drive your car looking in the rearview mirror

back2aukaatWhat you are saying is true for cyclical stocks. And this is Nifty PE and all the stocks in Nifty are not cyclical.

pattuPost authorI can't claim to know much about this but the movement of the Nifty is well correlated with its PE.

pattuPost authorReproducing my reply in AIFW, just for the record.

Every analysis is rearview mirror driving. I will agree that PE has issues but so have every metric!

In real-time it is difficult to use the PE but it has been done and we have at least one person here, Sundaramji, who has adopted this with reasonable success.

The correlation of market dips with PE (at least established indices like Nifty, Sensex) is high enough to warrant attention.

Anand DoctorShan, there is a solution for the point highlighted by you.There is another indicator that some of us use: PEG Ratio i.e. PE/Earnings Growth Rate.

This tells you if the price you're paying is justified by the earnings growth of the comapny/Index.

PEG = 1 indicates the index is Fairly priced.

PEG > 1 indicates the index is Expensive.

PEG < 1 indicates the Index is Cheap.

Hope this helps.

pattuPost authorThanks Anand. I played around with PEG sometime back and it simply too dependent on the earnings growth rate and most people simply assume that it is 15%

Nithin RAssuming other parameters being equal, between any two similar cap funds, how relevant is the data provided by valueresearchonline.com under portfolio characteristics section in download reports tab ? Can one take the PE ratio and divide it by 3Y growth (mentioned in the same section)to get the PEG ratio ? Will this simple analysis suffice to distinguish between two funds on PE/PEG basis ?

pattuPost authorI would suggest that you do not take the PE value of a fund seriously. Unless you know why the fund manager is holding a stock, the PE of a funds portfolio is not of much use imo.

AparnaPattu,

If you get so restless about getting rid of money from bank account, there is option to put it in liquid funds with 0% exit load right? Then it is also easy to switch or do STP when market PE reduces. I too get restless with money lying in account, because it is easily visible. I am planning to accumulate surplus in liquid funds.

I think the midway approach you suggested is good. Invest partially in SIP and hold the rest.

Aparna

pattuPost authorYes that is a good way to do it. Thought about it will see if I can do this too.

bharat shahone more benefit to keep in liquid funds rather than bank sb account apart from a little more earning , it is safer than bank in view of unauthorized withdrawal bank a/c by ATM or other ways , which is not possible for mutual fund. it is better to keep only month or two expenses in bank account, i think.

pattuPost authorGood points. Thank you.

JDso is it ok with the surplus of in 5 digits. Simple way , liquid funds are safe for even for longer period?

I am doing partly same way but still did not get confidence or say hesitating by keeping all in liquid.

Thanks

bharat shahby 'safe' i mean , it is not unauthorized with-drawable through misuse of ATM or other online ways unlike bank account , as the sale proceeds even by unauthorized would first deposit to your bank account , not to any other's bank account. and by the time you will get information of such unauthorized sale/redeem through your mobile /email, if you properly registered with your mf company before it is deposited to your bank account. you can take corrective actions.

k-dot-sheth-at-the-rate-gmail-dot-comWould love it if you could make the following simulation for indian stocks and share with your audience.

http://www.alphaarchitect.com/blog/2014/07/01/never-buy-expensive-stocks-period/

pattuPost authorWill try to. Thanks

Ashal JauhariHigh PE may be a sell sign but in case of a secular bull run and backed by good economy, the things may back fire.

Thanks

Ashal

pattuPost authorYes Ashal. Thanks is a risk to be borne by the investor.

Pattabiraman MurariThanks. You can get it from here: http://nseindia.com/products/content/equities/indices/historical_pepb.htm

AshokThanks Pattu for the extremely useful analysis - has just been wondering how far through the increasing PE should I keep investing.

For longer time frames (>15 years), even the modest returns(from a PE of around 25) seem to comfortably beat the FD returns post tax. So this warrants creating a basic, manual or automatic, SIP of around 50% to 75% of the monthly invest-able amount. Remaining money can be grown in a liquid fund until a lower PE opportunity presents itself.

pattuPost authorYes that is a very good observation. Thanks.

PARESHits good to know lots of funda to be seen in market. i do technical trading more. but this will help to get insight of state of market. thanks for your article. THUMPS UP !!! love your post ....... read all thru top 2 bottom...... scan like charts...

pattuPost authorThank you.

bharat shahtimely post as nifty pe is above 20. for me the centre idea is

1.' As mentioned above, the case for not investing at high PE (~ 25) remains strong enough, whereas the case for investing only at low PE ( 25).

3.However, it also makes sense to get rid of the salary into an investment each month. So perhaps one can have a base SIP running and ‘time’ the additional amount according to the PE. Perhaps.

solace in investing in sensible diversified equity mfs is that they come out of equity during market euphoria as far as possible. and it helps.

bharat shahtimely post as nifty pe is above 20.

for me the centre idea is

1.’ As mentioned above, the case for not investing at high PE (~ 25) remains strong enough, whereas the case for investing only at low PE ( 25).

2.Is SI-PE worth it? well, yes and no.

Yes it makes sense not to invest when the PE is high. Yes it even makes sense to pull out a portion of the equity corpus when the PE is very high (> 25).

However, it also makes sense to get rid of the salary into an investment each month. So perhaps one can have a base SIP running and ‘time’ the additional amount according to the PE. Perhaps

solace in investing in sensible diversified equity mfs is that they come out of equity during market euphoria as far as possible into cash/debt according to their mandate ,so it helps..

pattuPost authorYes I completely agree with you. Thanks.

BThere are certain pe based mfs also...can u compare there returns with ur anslysis??

pattuPost authorPE based mf also invest in debt and can sometimes be fully in debt. I am very comfortable with this comparison but will give it a shot.

shankarrajuHi Pattu, Thanks for the interesting article. Delaying for 2 weeks to find a right time to invest the surplus, makes sense.

The statistics says PE > 25 is not good time to buy. Let's say if the PE reduces from 25, according to the plots, if we buy at PE <=20, I think we have more probability for getting good returns irrespective of # of years holding. Am I right?

pattuPost authorI think investing at greater than 25 will impact returns. For all other values it does not matter. So if you have lump sum you can invest right now. Because over a long term the returns can just about be anything!

JDNice article once again...

How if we approach the concept via kind of MF? I mean if i just continue my SIP in FTIndia PE ratio FoF?

Also since long i am waiting for detailed analysis on this fund by you specially. In between i found your article on PE . so asked.

Keep it up your fantastic blogs..

Regards

Jig

pattuPost authorIf you are using the PE fof, just continue investing. Otherwise dont do anything unless the PE is extremely high.

sundararajanThe correlation is stronger, but the number of data points have reduced significantly. Those who invested at PE levels of 25 and above, 12 years back would be sitting with a return of about 10%, while those who invested with Nifty PE less than 15 would have got about 15% or more.

Are these 10% or 15% returns per year or for total period of 12/15 years? I assume that these returns are for per year. Can you confirm? Thanks,

pattuPost authorIt is return for the total period. In a volatile instrument it can never be per year return

sundararajanWhat? Am I missing something here? If 10% or 15% is total return for 15 years, it is too low on ROI. Assume one invests Rs 1 lakh for 12/15 years and if total return is 10%, one will get Rs 1.12 after 12 years or 1.15 after 15 years. Is it right?. I might have understood something wrong here. If it is 10% or 15% average return per year for 12/15 years make sense. I understand it can never be exactly 12/15 every year. Please let me know. Thanks,

pattuPost author10% is not the total return. It is the average returns one would have obtained each year. What I meant was, it the average and not equal to the return obtained each year. Should have clarified better. Apologies.

SundararajanThanks for the clarifications. I can breath now. 10% average return per year is perfectly make sense compare to inflation and bank FD returns.

tooheyCan you please share the excel sheet used to create the scatter charts in this article? thanks

pattuPost authorSure. You can get it here

http://freefincal.com/wp-content/uploads/2014/07/Automated-Rolling-returns-calculator-1.xlsb

Dilip kumarI have seen the very interesting plots which give nice insight into the return analysis for long term investor. Could you please share the excel sheet which is flexible with even more numbers of years is flexible.

Look forward for your response.

pattuPost authorSure. You can get it here

http://freefincal.com/wp-content/uploads/2014/07/Automated-Rolling-returns-calculator-1.xlsb

Chandrasekar RamakrishnanAwesome Professor!! I've been reading your articles / posts. But this post will make me follow you. An eye -opener. Thanks for the good work!!

Chandrasekar RamakrishnanAwesome Professor!! I've been reading your articles / posts. But this post will make me follow you. An eye -opener. Thanks for the good work!!

Pattabiraman Murarithank you very much.

AnirudhHello Pattu Sir,

Just wanted a clarification.....1st Jan 1999 to 30th June 2014 is the data range.....

So the scatter plots for each of the years(1,3,5) is ending 30th June 2014 or starting

1st Jan'99 ?!

Regards

Anish Menonsir please provide the excel sheet

Rajkumar SubbarayanVery Interesting analysis 🙂

Sanoopalbeit late, I am catching up with the interesting posts on your website 🙂

good articulated article.

freefincalPost authorThank you

Sudhansureally great article..... informative,,,,,, thanks

Satyajit MahapatroHi Sir

Can you please share the excel behind the analysis. Would want to go thru it .

May be we when we say investing at <15 , gives us a range of returns from 20% to 60% , we can probably assign a number to it based on weights.