How does the financial market research industry work?

[luutzu]

Hoping to get an education in how the stock market information industry works. That is, where does the financial data/ratios from sites like Yahoo!, Google, your brokers when open an account... and more specifically, how does research companies like Clime's StockInValue, StockDoctor get their data from?

From my understanding of the system, there seem to only be 3 financial information providers around the world - ThomsonReuters, S&P (which owns MorningStar) and Bloomberg - i know they're bonds, just assumed it's equities as well.

Your brokers and all other information providers seem to be getting equities data and ratios from these 3 (or their subsidiaries)... Most of the free ones seem to get the ratios straight from them and just presented it differently, slightly differently...

The better ones seem to do some calculations on their database from the standard financial statements from the big guys and present it to their design, but the source data are pretty much the same.

Other databases seem to either provide a window into MorningStar and that's about it.


So from what i understand, there seem to be no original/first hand research from pretty much any information provider - it's all just slightly different presentation from either one or two providers.

Am i right?

 

[DJG]

So from what i understand, there seem to be no original/first hand research from pretty much any information provider - it's all just slightly different presentation from either one or two providers.

Probably because they're all free websites. If you pay for the research service then you'll get far more in depth research reports.

Additionally they most likely get their numbers from the raw numbers off the financial statements. As bare in mind, these are the figures each company itself brings out. I'd recommend learning how to read them properly. More so than your standard Yahoo or Google Finance numbers - which seem to differ depending on whom.

 

[DeepState]

So from what i understand, there seem to be no original/first hand research from pretty much any information provider - it's all just slightly different presentation from either one or two providers.

Am i right?

I am wondering what are you expecting?

Adjusted ratios are a matter of opinion and the information providers are just trying to provide you with straight reads. For example, if you want to bring financial leases back on the balance sheet...that's for you to do. Or, if you want to change the depreciation schedule...again, how can they do that for you? Are below-the-line figures actually one-off...etc.

Is that what you want? If so, the data providers and portals just aren't in the position to supply it. You either have to get it off brokers or fund managers, who are not going to supply data to the portals. Or, you need to do the work yourself.

 

[luutzu]

Probably because they're all free websites. If you pay for the research service then you'll get far more in depth research reports.

Additionally they most likely get their numbers from the raw numbers off the financial statements. As bare in mind, these are the figures each company itself brings out. I'd recommend learning how to read them properly. More so than your standard Yahoo or Google Finance numbers - which seem to differ depending on whom.

Was just wondering where the analysts from the research companies get their information from.

I was assuming that the free ones were yea, from one or two sources that does the basic calculation.

But the paid research service provider, where they actually write detailed reports and have pretty charts... i was assuming that they actually get their information from the original Annual report, make adjustments etc... then write and recommend.

Maybe the guys from the big banks do i don't know, but the mid-range guys seem to just get it from Reuters or such and go with it.

 

[luutzu]

I am wondering what are you expecting?

Adjusted ratios are a matter of opinion and the information providers are just trying to provide you with straight reads. For example, if you want to bring financial leases back on the balance sheet...that's for you to do. Or, if you want to change the depreciation schedule...again, how can they do that for you? Are below-the-line figures actually one-off...etc.

Is that what you want? If so, the data providers and portals just aren't in the position to supply it. You either have to get it off brokers or fund managers, who are not going to supply data to the portals. Or, you need to do the work yourself.

The Reuters ones i saw are pretty good. There's a missing financial statement but what important is changes in equity... it's pretty good.. they even provided depreciation charges in year 1, 2 etc. too.

I'm not trying to be critical, just surprised that just about all the financial information (as in knowledge provider, not data) seem to get their data from a couple of sources and those sources, while very good, have a standardised format for the financial statements... and from Reuter say, it seems the Income Statement just contain the headline figures and no detailed from the notes.

For all i know that's all anyone need... was just a bit surpised those guys don't just open the annual reports themselves and see if there's anything in the notes worth looking at rather than rely on filtered data and work with what they get shown.

How does a fund manager or the research assistant/jnr analyst go about their research? From these portals/databases or these and combo of their own research/adjustments.

Not saying it's necessary to know the statements or know the costs and value to the last thousands... just trying to learn something new once a week

 

[DeepState]

The Reuters ones i saw are pretty good. There's a missing financial statement but what important is changes in equity... it's pretty good.. they even provided depreciation charges in year 1, 2 etc. too.

I'm not trying to be critical, just surprised that just about all the financial information (as in knowledge provider, not data) seem to get their data from a couple of sources and those sources, while very good, have a standardised format for the financial statements... and from Reuter say, it seems the Income Statement just contain the headline figures and no detailed from the notes.

For all i know that's all anyone need... was just a bit surpised those guys don't just open the annual reports themselves and see if there's anything in the notes worth looking at rather than rely on filtered data and work with what they get shown.

How does a fund manager or the research assistant/jnr analyst go about their research? From these portals/databases or these and combo of their own research/adjustments.

Not saying it's necessary to know the statements or know the costs and value to the last thousands... just trying to learn something new once a week

Hi Luu

I'm learning too.

Raw data comes in from the providers in standard templates. This is required because comparisons must be facilitated and aggregations of data are also required to calculate things like market-wide ratios etc. It is always a trade off to squeeze what the company has written and reported into standard templates. Some data aggregators will allow you to see where the reported template figure came from in the annual reports. This allows you to see what the assumptions made were in that process.

No single fund manager or broker has the time to maintain this level of data input. It is handed to the data suppliers who report it to the aggregators whose data is then used by clients like fund managers, traders etc.

The professional market uses this data for aggregate market statistics and to screen up for interesting situations that may require further investigation. This aggregate data, and company level data, includes a line item for depreciation and amortization. In many cases, the annual reports or proxies for these are considered in depth which includes detailed reading and analysis of the notes to accounts. These make up nearly half or more of annual reports. So the market does its own work beyond the template reports. Not sure how you came to conclude they did not.

These adjustments are a matter of opinion. Two analysts can come to different perspectives on the true rate of depreciation on the stock of capital or the rate of amortization of goodwill, for example. The accounting standards are now very prescriptive and this can and does subtract from the ability to discern the true economic picture. Adjustments are made which, according to the analyst, helps provide an even truer and fairer picture of what is going on in the company. This can be further supplemented by industry analysis, meetings with the company CFO etc. to get an even better idea of what has been reported and the prospects for the company on an ongoing basis.

Cheers

 

[luutzu]

Thanks RY,

Yea was assuming the fund managers and analysts use these data, and with other research, to screen or get an overview then do more detailed work if it look interesting enough.

Was curious because I saw some research house/fund manager selling software and research covering hundreds of companies, one even cover thousands, and all they have in their team are four or five people... there's a team leader/director, this other one have a couple of developer... all in all maybe 2 to 3 analysts and the rest are sales or support.

Then the question becomes... from an operational/practical point of view, for the research houses and for the fund managers/analysts... by getting pretty much the same aggregate information as every one else in the industry, these even come with consensus estimates as well... how and when do they decide to take a closer look.

If say, the data on their screen say PE ratio = 20. If current price is $40, then earnings must be $2... but if the database get $2 EPS by averaging analysts' estimates EPS for next year or two.. and so you read earning is $2 and this industry is going down the tube, you probably lowered that consensus further or just ignore the company completely.

I think analysts would look at a few other factors and those would give context and paint a better picture.

I guess it is generally useful as long as you know how the data and the results were put together.


---

and when the analyst decide to take a closer look. What tools do they use to change the figures and make their own assumptions?

Unless their firm has that multi million dollar software, they probably use Excel or Access? Or just pen and paper?

 

[DeepState]

1. Thanks RY,

2. I saw some research house/fund manager selling software and research covering hundreds of companies, one even cover thousands, and all they have in their team are four or five people... there's a team leader/director, this other one have a couple of developer... all in all maybe 2 to 3 analysts and the rest are sales or support.

3. Then the question becomes... from an operational/practical point of view, for the research houses and for the fund managers/analysts... by getting pretty much the same aggregate information as every one else in the industry, these even come with consensus estimates as well... how and when do they decide to take a closer look.

4. When the analyst decide to take a closer look. What tools do they use to change the figures and make their own assumptions? Unless their firm has that multi million dollar software, they probably use Excel or Access? Or just pen and paper?

1. No probs. Welcome.

2. Developing databases and updating figures for 1,000 firms who produce HY and FY results, let alone quarterly production is a monster task. That's 2,000 3-way accounts at least per year. If each takes, say, 4 hrs because they are trying to add some value somehow (and that is no time at all to try and figure out what is going on and recasting historical earnings), we are looking at 8,000 person hours per year. Each person, working 8 hours per day for 47 weeks a year produces 1,880 person hours. They may work longer hours but I have allowed zero time for meetings, toilet stops, smoking breaks, illness...in that calculation.

A team of that size has no chance to be an original producer of current data, let along have the time to backfill a decade worth of accounts, let alone know anything of the company outside of reading about them in the company accounts (they would not have time to read the report itself) sufficient to make a recommendation which could credibly be better than a spin on Twister. It takes years to understand what is happening and a typical fund manager of, say, six people, is often allocated two weeks per stock just to keep up with what is happening to a degree sufficient to have a reasonable basis to make the recommendation.

They are using aggregated data of some sort. If they are seeking to add value to that via in-house research etc. Good luck.

3. This type of data is not where the competition is played out. Everyone has it - as you say. It's not about getting a better data feed for historical accounts. It is in how they are interpreted, how earnings are stabilised, how assets are valued, understanding how cash moves around etc..that gives value to them. Then, you use these as one part of your process of projections which then feeds into your estimate of intrinsic value or helps figure out where the next delta (change in something) is if you care about short-term prediction. Everything you can get your hands on is fair game for use in projections.

Each analyst has their own idea of what they are looking for. There is no rule that is universal. Some look for low P/E, some look for high growth etc.. Any stock which is expected to move up faster than your stated objective is a cheap stock, even if it trades at a P/E of 20 or more. A P/E of 5x doesn't make a stock cheap either. Some hunt in junk looking for a turnaround, others will only invest in companies that have high-quality earnings. It all depends. There is no correct answer, no formula for success.

4. When they do their analysis, a spreadsheet is usually sufficient for a fundamental analyst.

 

[luutzu]

1. No probs. Welcome.

2. Developing databases and updating figures for 1,000 firms who produce HY and FY results, let alone quarterly production is a monster task. That's 2,000 3-way accounts at least per year. If each takes, say, 4 hrs because they are trying to add some value somehow (and that is no time at all to try and figure out what is going on and recasting historical earnings), we are looking at 8,000 person hours per year. Each person, working 8 hours per day for 47 weeks a year produces 1,880 person hours. They may work longer hours but I have allowed zero time for meetings, toilet stops, smoking breaks, illness...in that calculation.

A team of that size has no chance to be an original producer of current data, let along have the time to backfill a decade worth of accounts, let alone know anything of the company outside of reading about them in the company accounts (they would not have time to read the report itself) sufficient to make a recommendation which could credibly be better than a spin on Twister. It takes years to understand what is happening and a typical fund manager of, say, six people, is often allocated two weeks per stock just to keep up with what is happening to a degree sufficient to have a reasonable basis to make the recommendation.

They are using aggregated data of some sort. If they are seeking to add value to that via in-house research etc. Good luck.

3. This type of data is not where the competition is played out. Everyone has it - as you say. It's not about getting a better data feed for historical accounts. It is in how they are interpreted, how earnings are stabilised, how assets are valued, understanding how cash moves around etc..that gives value to them. Then, you use these as one part of your process of projections which then feeds into your estimate of intrinsic value or helps figure out where the next delta (change in something) is if you care about short-term prediction. Everything you can get your hands on is fair game for use in projections.

Each analyst has their own idea of what they are looking for. There is no rule that is universal. Some look for low P/E, some look for high growth etc.. Any stock which is expected to move up faster than your stated objective is a cheap stock, even if it trades at a P/E of 20 or more. A P/E of 5x doesn't make a stock cheap either. Some hunt in junk looking for a turnaround, others will only invest in companies that have high-quality earnings. It all depends. There is no correct answer, no formula for success.

4. When they do their analysis, a spreadsheet is usually sufficient for a fundamental analyst.

I read "Masters of the Market" a while back. The fund managers interviewed are, as the authors claimed, among the best in Australia, each with a long record of outstanding performances, managing hundreds of millions and billions.
Most starts out not really knowing what they were doing, and after losing god know how much of other people's money, soon learn and got better..

They sounds like reasonably decent people, but it makes you wonder about the other smart monies who's watching over our trillions [?] in retirement savings.

Are you saying that they're getting information from these providers and if it looks interesting, do a few calculations on excel?

---
There is a formula, just the formula is "it depends". That make it appear like there's no formula or logic to follow, but there is, at least i think so anyway.

So if high level of cash, improved profit margin from previous reporting period... is good or bad depends on something, how will an analyst who spent 10 days on the stock know what that dependencies are to make a reasonable estimate of new events on the company? Just guess work then.

Like the PE example. A high PE might not mean overpriced as a low PE doesn't mean cheap. Cheap or not depends on the eventual E, not on whether the analysts all got the correct estimate of E but some prefer a higher PE ratio while others prefer cheaper stocks so go for lower PE. But this tend to happen, from what i can tell anyway, and it's a bit weird.

I think Shiller did some research and found at in general, over the last century, buying lower PE stocks actually return better than buying higher PE stock over a given 10 year period

---

RY, mine sharing with us what kind of system your team was building back then? The one you said you have a baby version of.

 

[DeepState]

1. I read "Masters of the Market" a while back. The fund managers interviewed are, as the authors claimed, among the best in Australia, each with a long record of outstanding performances, managing hundreds of millions and billions. Most starts out not really knowing what they were doing, and after losing god know how much of other people's money, soon learn and got better..

They sounds like reasonably decent people, but it makes you wonder about the other smart monies who's watching over our trillions [?] in retirement savings.

2. Are you saying that they're getting information from these providers and if it looks interesting, do a few calculations on excel?

---
3. There is a formula, just the formula is "it depends". That make it appear like there's no formula or logic to follow, but there is, at least i think so anyway.

So if high level of cash, improved profit margin from previous reporting period... is good or bad depends on something, how will an analyst who spent 10 days on the stock know what that dependencies are to make a reasonable estimate of new events on the company? Just guess work then.

Like the PE example. A high PE might not mean overpriced as a low PE doesn't mean cheap. Cheap or not depends on the eventual E, not on whether the analysts all got the correct estimate of E but some prefer a higher PE ratio while others prefer cheaper stocks so go for lower PE. But this tend to happen, from what i can tell anyway, and it's a bit weird.

4. I think Shiller did some research and found at in general, over the last century, buying lower PE stocks actually return better than buying higher PE stock over a given 10 year period

---

5. RY, mine sharing with us what kind of system your team was building back then? The one you said you have a baby version of.

1. When you start out, they don't hand you the reigns to a billion dollar portfolio. As with other apprenticeships, you start by doing the grunt work. You make no decisions that affect portfolios. Over time, as you show ability, you get more responsibility.

Even amongst those giants, as published in 2005, it is interesting to see where they went. Robert-Maple Brown's eponymous firm has fallen on hard times and their last five year results are in line with peers. Peter Morgan was awesome from a young age and Perpetual and was helped along by a guy called Peter Rayner to become head of equities there and then formed his own firm. He did well, but pulled out after a cancer scare (false positive) and got himself a life. Greg Perry retired not long after to spend time with his autistic son and tend to his horses, but he was an amazing investor with a photographic memory. Anton (whose phone anwering technique is memorable amongst his ex-colleagues at County) left County Natwest as a mid-tier analyst and somehow built an incredible firm in Investors Mutual which is still doing well. Tim Hughes lost his job at Rothschild and his private fixed income firm that was formed shortly afterwards with Melissa Ayton closed shortly afterwards due to poor investment performance. After a time at Catholic Superannuation Fund as CIO, he is now CIO at Non-Government Schools Superannuation Fund. These roles are not fund manager roles although they are still investment related. I don't know about Wilson or Brierly first hand as Wilson is mostly retail and Sir Ron operated from within a listed entity as a raider. Paradice is an awesome guy and has built up Paradise Asset Management into a giant and continues to post decent numbers.

So the failure rate is still reasonably high even amongst these super-stars. Or other life priorities jump in the way as well.



2. Yep. But each person's analysis will be different to the next. That is the difference. And those numbers are considered with knowledge of the context of the company and its circumstances. It's not just rehashing some figures on a 3-way account in isolation. They understand what is happening and try to express that in numbers. Funny story, in about 1995, we visited Maple Brown Abbott. At this time, everyone had a PC on their desks and were whirring away. MB-A had one single PC in the whole firm. It was not even switched on when we were there. Yet they did really well for a time. Pen and paper. Some people don't do much of this stuff at all. They know it boils down to a few key variables and that valuation ranges are so wide it is not worth bothering. Lots of things drive the market. Historical accounts are just one.


3. You are not going to find a book that seriously and accurately gives to the 10 step program to becoming rich via investment. You might find biographies but nothing that really makes you rich. Notice how there are no technical traders in that list? Zero for Eight. Lots of books on bookshelves telling you how to chart though. None of the Masters have written a book either. Each of them is very different to the next. There is no formula. Even they couldn't lay out a 10-step program of why they are successful. It is judgment. It depends. They are just really really good at figuring out what it depends on in the particular circumstance. They have innate understanding of markets and they honed this with experience. And...even then...it may not be enough.

An analyst may be budgeted to spend 10 days per stock - on average - in their coverage universe. With experience, you know what to look for and what sources of uncertainty you must cover. Investigations take as long as is necessary to develop conviction acknowledging that conviction can not lead to certainty. You are looking for an edge.

4. This is the value effect. Maple Brown-Abbott is a value investor. If you like, you can do this. But others who got wealthy as well did not tilt this way because they were able to earn more from the markets than what could be earned from investing in the value phenomenon. However, it could be part of how you invest. That is for you to decide.

5. Please refer to Post #25 of Trading Strategies / Systems...
"Backtesting based on Fundamental Data"


There is no system, formula, data source or fixed way to make money. It boils down to the superior judgments of the investors and their initiative to develop an edge of some sort. Success comes in many colours and so does failure. You just have to figure it out for yourself and learn. The last place you should learn from is Uni. Personally I was so pleased that each year thousands of new kids would be freshly minted finance graduates and hit the market with all sorts of crazy notions which they were taught and scored H1 in. None of the Masters have a PhD. But nothing prevents you from getting good ideas of how to go about it via studying history and seeking to understand how great investors think as well.

A good place to start is finding out about what Charlie Munger says are the Big Ideas. Learn about those and you might improve your chances of success.

 

[luutzu]

Appreciate the time RY, I get a bit smarter each time i talk to you.

From the interviews, Paradice, Perry and Morgan sounds like genuinely decent people, the others i can't say, not that i can of the three from a couple of interviews.

I don't believe in innate ability or natural born anything. People can learn and master anything they set their mind to.
That's not a cliche, it's what i've seen.

I've worked with very smart, very experienced people and, from what i see, they don't make good decisions because they're gifted or just somehow brilliant. They make good decisions when given good information, or know how to get the right information. The same genius have made bad decisions when the info are bad just as an average person could make the right call when the data is pretty obvious.

Sometimes you have to make judgments from incomplete information... but as Buffett have said, you're the guy with the money, if an opportunity is uncertain, there's no need to act - just wait for the next one you're sure of.

You're right that there's no formula, no prescriptions a person can follow. I don't think there could ever be one. Academics have and are trying to get to it but it's mostly rubbish. I find some general thesis behind the F or Z scores are sound, but to boil it down to a number is insane. It doesn't make sense to try to precisely predict the future.

-----


I think people in general got the wrong impression that Graham is all about value. From the little i read about him, value (as in cheap) stock is what he look for... but his teachings recognise that riches are better found in future growth from a great business, not in net asset value bargains or low PE stocks... just i think he doesn't want to pay for or risk a bigger future payoff when he could get a still reasonable payoff right now with certainty.

When we think about it, I don't think there is such thing as a value vs growth investor. If a person is an investor, he is always a value investor, even when he invest in "growth" companies.

When a person buy what looks like a low PE stock, it's not, or shouldn't be, done because he expect the company to grow its earnings less than if he buy a high PE stock. The expectations of growth, as in return on his investments, are the same... just with the low PE, he's getting a better bargain than if the asking price is higher.

The other reasons for his demanding a lower price, so lower PE, would be due to possibility of minor loss or less timely earnings growth... and his willingness to pay at what appear to be a slightly higher PE ratio should be because of greater certainty of his judgment, or top quality assets and management.

basically, the investor should always work within a range of value... not either growth and pay 40 or 50 times earnings and never 5 or 10 times earnings.

---

I actually quit my Masters at NSW because one of the courses taught advanced CAPM and other spreadsheet modelings in investment. Asking me to pay $5000 on something so wrong and useless.

I find Buffett and Munger quite brilliant. And Buffett is right that there's no use writing another investment book because Graham has said and done it so well very few could hope to surpass them. Maybe with the exception of Phillip Fisher.

But I think their advice has been caricature-ized though. It's become the "Buffett Way" and so on.

The approach is simple and doable, just the practice is not practicable for most people who has a day job other than business analysis, and whose cost/benefit equation meant they can't work on it long enough to be good and always be ready for opportunities when it come.

 

[DeepState]

I don't believe in...

Ultimately, investment is a contest of beliefs, temperament and opportunity. You have a lot of beliefs you hold to be true for whatever reason. You have the opportunity to enter the market, although the market may not offer the opportunities for your beliefs to play through at any given time. Your temperament is unique to you.

You'll need to put them all to the test and see what happens.

 

[luutzu]

Ultimately, investment is a contest of beliefs, temperament and opportunity. You have a lot of beliefs you hold to be true for whatever reason. You have the opportunity to enter the market, although the market may not offer the opportunities for your beliefs to play through at any given time. Your temperament is unique to you.

You'll need to put them all to the test and see what happens.

yea, beliefs can be a dangerous thing

I find it strange that a master plumber would never be considered innately talented at plumbing but a master fund manager would be considered somehow wired or made for reading the market, human psychology and the financial statements just sings to him.

Or that a rennovator who take seconds to know where the toilet, the shower, the vanity, the plumbing, the tile patterns... should go... then set about welding the water pipes, running the sewage, waterproof, tile, fix the lights and power... that guy is just a jack of all trade but a richer guy who have learnt to read business news, economic conditions, annual reports and maybe write a poem or two.... that guy is a Renaissance man.

You wouldn't find a blue collar guy ever claiming he's good at what he does because he's just gifted, but there's plenty of rich professionals that think they're made for the job and paid plenty of cash because of inherent talent and natural ability.

I met this guy who worked for some investment bank. The bank flew him to Hong Kong to give some trustees or major investors assurance of the soundness of this and that product and fund... and he should know because why else would they flew him all the way to HK. The guy didn't know anything about what he's selling, and he's not from the sales department either... he learn key figures and product details from the brochures the bank provide, while he's on the plane.

If a plumber doesn't know what he's doing, the toilet might be blocked or the sewage might run down the street... financial advisors and investment professional not knowing and people could lose their retirement and life's savings.