Dow Jones Earnings and Debt Since 2011

[Guest Cameron]

Spent my Saturday night looking at the Dow Jones as a whole you can see with the excel sheet attached.

 

Second sheet that looks at debt has a row at the bottom that subtracts the banks debt and earnings as well as GE because of GE Capital, with GE and the Banks debt doesn't really look like it grew over a 5 year period but its clear it grew fast for the other 27 companies.

 

Earnings also peaked in 2013 and have been down 7% since 2011 yet the Dow has returned 90%.

 

Going to add a couple more spreadsheets that show EPS, share buybacks etc.

 

I'm thinking about maybe doing something similar with the S&P 500 but the shear amount of companies makes it seem unrealistic.

ExchangeAnalysis.xlsx

US_Macro.xlsx

[Uccmal]

Nice work. 

 

It seems to tell us that the majority of the Dow Companies are gorging on debt.  We know that 3 or 4 have huge cash balances. 

 

I think you would see the same rising debt levels across the entire S&P.  I have also seen charts showing the EPS for the S&P which peaked a few years ago. 

 

To me this all illustrates a big problem for central bankers worldwide.  They want to raise interest rates but I suspect they wont be able to.  In fact I would bet that as soon as this shows signs of cracking, interest rates will come down again. 

 

The very best scenario is likely extremely low stock returns going forward for the indexes.  But things never really work that way.  We are more likely to get a significant correction. 

[Guest Cameron]

Updated with Book Value and Free Cash Flow

 

DowJonesAnalysis_2.xlsx

[shalab]

Great work - if one looks carefully, Verizon, Microsoft and apple are the main culprits. In Microsoft's case, net income and stock price have moved in opposite directions

[rb]

The general idea is correct, but it's not as dramatic as the OP suggests.

 

When talking about earnings there's a commodities story at play where Exxon, Chevron and CAT got smacked hard. Adjust for those and what you get is that earnings peaked in 2014, dipped in 2015 and recovered in 2016 to 2014 levels. Taking out the commodities story you get earnings growth on 2011-2014 of 20% as opposed to -7%. If you pay attention to AAPL, you notice that both the earnings dip in 2015 and recovery in 2016 were quite pronounced.

 

Yes the stock prices did very well. But it's not a secret that the past few years have been very good for stocks.

 

On the debt side once you make a few adjustments the idea is the same but the magnitude is less. The OP rightfully excluded banks and GE which would distort the picture. I also adjusted for MSFT and AAPL for whom this is not really leverage but rather tax strategy. Once you make those adjustments you get debt going from $495B in 2011 to $762B in 2016. And increase of $267B. Out of that $53B was Verizon who did it to buy 50% of Verizon Wireless so let's call that good debt. On the flip side Chevron and Exxon borrowed $62B extra likely driven by the hard times they're facing.... so i guess that's bad debt.

 

Between Chevron, Exxon and Verizon that's 43% of the total debt increase. So if you account for a few special cases and stories again the magnitude decreases. Nonetheless the premise stands that these companies levered up over the past 5 years. A bit too much fun with the buybacks I think. The debt levels are still totally manageable for these sort of companies but probably less buybacks in the future.

 

Bottom line? Stocks are kinda pricey now. I do want to thank the OP for putting in the grunt work of getting all this data together. It's helpful to have it all to play around to do big picture stuff.

[Guest Cameron]

The general idea is correct, but it's not as dramatic as the OP suggests.

 

When talking about earnings there's a commodities story at play where Exxon, Chevron and CAT got smacked hard. Adjust for those and what you get is that earnings peaked in 2014, dipped in 2015 and recovered in 2016 to 2014 levels. Taking out the commodities story you get earnings growth on 2011-2014 of 20% as opposed to -7%. If you pay attention to AAPL, you notice that both the earnings dip in 2015 and recovery in 2016 were quite pronounced.

 

Yes the stock prices did very well. But it's not a secret that the past few years have been very good for stocks.

 

On the debt side once you make a few adjustments the idea is the same but the magnitude is less. The OP rightfully excluded banks and GE which would distort the picture. I also adjusted for MSFT and AAPL for whom this is not really leverage but rather tax strategy. Once you make those adjustments you get debt going from $495B in 2011 to $762B in 2016. And increase of $267B. Out of that $53B was Verizon who did it to buy 50% of Verizon Wireless so let's call that good debt. On the flip side Chevron and Exxon borrowed $62B extra likely driven by the hard times they're facing.... so i guess that's bad debt.

 

Between Chevron, Exxon and Verizon that's 43% of the total debt increase. So if you account for a few special cases and stories again the magnitude decreases. Nonetheless the premise stands that these companies levered up over the past 5 years. A bit too much fun with the buybacks I think. The debt levels are still totally manageable for these sort of companies but probably less buybacks in the future.

 

Bottom line? Stocks are kinda pricey now. I do want to thank the OP for putting in the grunt work of getting all this data together. It's helpful to have it all to play around to do big picture stuff.

 

No problem, I added the attachment below and made the adjustment to the net income sheet to exclude CAT, XOM, and CVX, I made it late at night and completely forgot about oil falling in 2014. The picture looks definitely different when excluding those 3, I also added EPS and Revenue, I'll be trying to add another 5-10 years to the data as well and have started on the S&P 500.

 

Most interesting sheet to me is book value and the ROE. Only other time Book Value was near 360% was Black Monday, the tech bubble, before the housing crisis and now. But then again treasuries and interest rates weren't as low as they are today.

[Valuehalla]

THX for the great work Cameron.

 

To my mind its just important what are the future expectations concerning earnings & FCF of these companies. Than you can think about the relations to the current stockprice/marketcap and than make a decision, if a buy is a good investment in your individual situation or not. One will love 5 % return, another wants to see minimum 15 % etc..

 

The past is not relevant.

 

In total, as I remember, WEB said in the latest interviews, that pricevaluations in avg of the S&P 500 are now mediocre in historic comparisons.

 

I have the following historical figures to contribute as mosaic pieces:

- The average PE ratio of S&P 500 from 1936 to 2015 is 16.9

- From 1965 to 2016 the S&P 500 gained in avg 9.7 % incl. dividends

- Buying power of 0.13 US$ in 1965 is like 1 US$ in 2015, which is an inflation of 4,4 % p.a. avg 50 years

 

If one of my figures is not correct, let me know.

 

 

[Guest Cameron]

THX for the great work Cameron.

 

To my mind its just important what are the future expectations concerning earnings & FCF of these companies. Than you can think about the relations to the current stockprice/marketcap and than make a decision, if a buy is a good investment in your individual situation or not.

 

One will love 5 % return, another wants to see minimum 15 % etc..

 

The past is not relevant.

 

In total picture, as I remember, WEB said in the latest interviews, that pricevaluations in avg of the S&P 500are now mediocre in historic comparisons.

 

I have the following historical figures:

- The average PE ratio of S&P 500 from 1936 to 2015 is 16.9

- From 1965 to2016 the S&P 500 gained in avg 9.7 % incl. dividens

- Buying power of 0.13 US$ in 1965 is like 1 US$ in 2015, which is a inflation of 4,4 % p.a. avg 50 years

 

If one of my figures is not correct, let me know.

 

Current PE of the S&P is around 24x so I'll have to find the interview to see the context of what he meant by mediocre considering its higher than the norm I'm assuming he meant returns at this level.

 

I don't have the S&P 500 historical returns but I added from 2004-2016 net income in the sheet as well as the historical 1 year 5 year 10 year and 20 year returns of the Dow Jones since 1896 if that interests anyone

 

 

[Valuehalla]

Cameron i downloaded you latest updated file.

 

Could you maybe post a keypoint-list with your main conclusions which result in your opinion from your statistic.

[rb]

THX for the great work Cameron.

 

To my mind its just important what are the future expectations concerning earnings & FCF of these companies. Than you can think about the relations to the current stockprice/marketcap and than make a decision, if a buy is a good investment in your individual situation or not. One will love 5 % return, another wants to see minimum 15 % etc..

 

The past is not relevant.

 

In total, as I remember, WEB said in the latest interviews, that pricevaluations in avg of the S&P 500 are now mediocre in historic comparisons.

 

I have the following historical figures to contribute as mosaic pieces:

- The average PE ratio of S&P 500 from 1936 to 2015 is 16.9

- From 1965 to 2016 the S&P 500 gained in avg 9.7 % incl. dividends

- Buying power of 0.13 US$ in 1965 is like 1 US$ in 2015, which is an inflation of 4,4 % p.a. avg 50 years

 

If one of my figures is not correct, let me know.

I don't want to nitpick case I don't mean anything by it.

 

About your figures.... I don't have a good data source about total stock market return (if anyone does please share). About inflation though, you're right about the 13 cents (not that that means anything). However that's 4.1% inflation not 4.4% - hey, you asked.

 

Now that includes the 70s which had bad inflation and bad stock market. Just like anything involving statistics your start and end points matter a lot. if we were to look at the past 30 years ('85-'15) as opposed to the past 50, inflation becomes 2.7% and I think stock market performance becomes better too. I guess the biggest question is what will the next 30, 50 years or whatever look like? Is it like the past 30, the past 50, or something else alltogether? I don't think anyone can tell for sure.

[TwoCitiesCapital]

The general idea is correct, but it's not as dramatic as the OP suggests.

 

When talking about earnings there's a commodities story at play where Exxon, Chevron and CAT got smacked hard. Adjust for those and what you get is that earnings peaked in 2014, dipped in 2015 and recovered in 2016 to 2014 levels. Taking out the commodities story you get earnings growth on 2011-2014 of 20% as opposed to -7%. If you pay attention to AAPL, you notice that both the earnings dip in 2015 and recovery in 2016 were quite pronounced.

 

Yes the stock prices did very well. But it's not a secret that the past few years have been very good for stocks.

 

On the debt side once you make a few adjustments the idea is the same but the magnitude is less. The OP rightfully excluded banks and GE which would distort the picture. I also adjusted for MSFT and AAPL for whom this is not really leverage but rather tax strategy. Once you make those adjustments you get debt going from $495B in 2011 to $762B in 2016. And increase of $267B. Out of that $53B was Verizon who did it to buy 50% of Verizon Wireless so let's call that good debt. On the flip side Chevron and Exxon borrowed $62B extra likely driven by the hard times they're facing.... so i guess that's bad debt.

 

Between Chevron, Exxon and Verizon that's 43% of the total debt increase. So if you account for a few special cases and stories again the magnitude decreases. Nonetheless the premise stands that these companies levered up over the past 5 years. A bit too much fun with the buybacks I think. The debt levels are still totally manageable for these sort of companies but probably less buybacks in the future.

 

Bottom line? Stocks are kinda pricey now. I do want to thank the OP for putting in the grunt work of getting all this data together. It's helpful to have it all to play around to do big picture stuff.

 

What's the logic of defining good debt/bad debt tax strategy/leverage? Debt is debt. It's good when prices/profits/revenues are rising. It's bad when they're falling. All that matters is how an additional dollar of debt impacts the incremental net profits of a company - in good times that will be a positive. In bad times it will be a negative - the purpose/strategy of why/how the debt was added doesn't really matter when it comes due and needs to be paid or rolled at an inopportune time.

 

 

 

 

 

[rb]

What's the logic of defining good debt/bad debt tax strategy/leverage? Debt is debt. It's good when prices/profits/revenues are rising. It's bad when they're falling. All that matters is how an additional dollar of debt impacts the incremental net profits of a company - in good times that will be a positive. In bad times it will be a negative - the purpose/strategy of why/how the debt was added doesn't really matter when it comes due and needs to be paid or rolled at an inopportune time.

The logic is that nuance of how the debt comes to be. There's a world of difference between MSFT and AAPL and CVX and XOM. MSFT and AAPL are taking on debt to lower their taxes and have their debt covered 2x by cash in their bank accounts. CVX and XOM are taking on debt to pay their bills because their profits just cratered. If the inopportune moment as you call it comes MSFT and AAPL can just pay a bit of tax and extinguish their debt but XOM and CVX are stuffed. Huge difference.

 

We're looking at these data sets in order to draw some informed conclusions. Adjusting the data with information we have about special situation improves the data and leads to better conclusions.

[TwoCitiesCapital]

What's the logic of defining good debt/bad debt tax strategy/leverage? Debt is debt. It's good when prices/profits/revenues are rising. It's bad when they're falling. All that matters is how an additional dollar of debt impacts the incremental net profits of a company - in good times that will be a positive. In bad times it will be a negative - the purpose/strategy of why/how the debt was added doesn't really matter when it comes due and needs to be paid or rolled at an inopportune time.

The logic is that nuance of how the debt comes to be. There's a world of difference between MSFT and AAPL and CVX and XOM. MSFT and AAPL are taking on debt to lower their taxes and have their debt covered 2x by cash in their bank accounts. CVX and XOM are taking on debt to pay their bills because their profits just cratered. If the inopportune moment as you call it comes MSFT and AAPL can just pay a bit of tax and extinguish their debt but XOM and CVX are stuffed. Huge difference.

 

We're looking at these data sets in order to draw some informed conclusions. Adjusting the data with information we have about special situation improves the data and leads to better conclusions.

 

But we're using it to inform us about the aggregate index as a whole. Sure, if we want to know about MSFT's prospects, it might make sense to adjust, but if we're looking at it to inform us on the aggregate of business/finances conducted by the DJIA, both debts contribute and both cash balances (or lack thereof) also contribute. It doesn't make sense to me remove "probablamatic" data points to get an idea of the whole unless if those datapoints are wild extremes that skew results. Arguably, it's not that wild for corporatations to use debt in tax planning strategies and I don't see why they should be excluded from index level statistics noting the increase in indebtedness of large corps in the U.S.

 

That being said, we both would probably agree that median statistics are more meaningful here. What has happened to the debt/EBITDA ratio of the meadian company in the DJIA/S&P 500? How has the statistics of that median level and the financials of each year's median evolved and you probably have a more clear picture of what is going on with corporate America's finances.

[rb]

That being said, we both would probably agree that median statistics are more meaningful here. What has happened to the debt/EBITDA ratio of the meadian company in the DJIA/S&P 500? How has the statistics of that median level and the financials of each year's median evolved and you probably have a more clear picture of what is going on with corporate America's finances.

 

I think i covered that in my post:

 

Between Chevron, Exxon and Verizon that's 43% of the total debt increase. So if you account for a few special cases and stories again the magnitude decreases. Nonetheless the premise stands that these companies levered up over the past 5 years. A bit too much fun with the buybacks I think. The debt levels are still totally manageable for these sort of companies but probably less buybacks in the future.

 

Bottom line? Stocks are kinda pricey now. I do want to thank the OP for putting in the grunt work of getting all this data together. It's helpful to have it all to play around to do big picture stuff.

[TwoCitiesCapital]

That being said, we both would probably agree that median statistics are more meaningful here. What has happened to the debt/EBITDA ratio of the meadian company in the DJIA/S&P 500? How has the statistics of that median level and the financials of each year's median evolved and you probably have a more clear picture of what is going on with corporate America's finances.

 

I think i covered that in my post:

 

Between Chevron, Exxon and Verizon that's 43% of the total debt increase. So if you account for a few special cases and stories again the magnitude decreases. Nonetheless the premise stands that these companies levered up over the past 5 years. A bit too much fun with the buybacks I think. The debt levels are still totally manageable for these sort of companies but probably less buybacks in the future.

 

Bottom line? Stocks are kinda pricey now. I do want to thank the OP for putting in the grunt work of getting all this data together. It's helpful to have it all to play around to do big picture stuff.

 

Sure - it wasn't so much the conclusion I was probing as much as the thought process on differentiation debts of individual companies for aggregate statistics.

[rukawa]

The logic is that nuance of how the debt comes to be. There's a world of difference between MSFT and AAPL and CVX and XOM. MSFT and AAPL are taking on debt to lower their taxes and have their debt covered 2x by cash in their bank accounts.

 

Then it would make sense to look at something like enterprise value which adds the debt to the market cap and subtracts out the cash or failing that at least subtract cash from the total debt. I am as suspicious as TwoCitiesCapital of selectively removing individual companies.

 

I'm thinking about maybe doing something similar with the S&P 500 but the shear amount of companies makes it seem unrealistic.

 

I could probably get this off of Bloomberg or failing that we could use this:

https://github.com/JeffFerguson/secqdb

 

I am going to try and see what I can grab off of secqdb...even though that is the harder route...because I've basically been itching to see how far the structured datasets can go. But if I can do it by the end of the week...Bloomberg it is.

 

[Guest Cameron]

The logic is that nuance of how the debt comes to be. There's a world of difference between MSFT and AAPL and CVX and XOM. MSFT and AAPL are taking on debt to lower their taxes and have their debt covered 2x by cash in their bank accounts.

 

Then it would make sense to look at something like enterprise value which adds the debt to the market cap and subtracts out the cash or failing that at least subtract cash from the total debt. I am as suspicious as TwoCitiesCapital of selectively removing individual companies.

 

I'm thinking about maybe doing something similar with the S&P 500 but the shear amount of companies makes it seem unrealistic.

 

I could probably get this off of Bloomberg or failing that we could use this:

https://github.com/JeffFerguson/secqdb

 

I am going to try and see what I can grab off of secqdb...even though that is the harder route...because I've basically been itching to see how far the structured datasets can go. But if I can do it by the end of the week...Bloomberg it is.

 

As a result I am working on adding a net cash sheet to the document, I have updated the doc with historical NASDAQ returns, results are pretty interesting, I'm going to add as many exchanges with as much data I can get my hands on

 

If anyone has any other suggestions I'm game. I'm trying to see if there is anyway I can get full book value and market caps for every stock in an exchange to pin point cheap exchanges relative to each other as well.

[rb]

The logic is that nuance of how the debt comes to be. There's a world of difference between MSFT and AAPL and CVX and XOM. MSFT and AAPL are taking on debt to lower their taxes and have their debt covered 2x by cash in their bank accounts.

Then it would make sense to look at something like enterprise value which adds the debt to the market cap and subtracts out the cash or failing that at least subtract cash from the total debt. I am as suspicious as TwoCitiesCapital of selectively removing individual companies.

 

Of course working with net debt numbers makes a lot more sense. But I was pretty lazy yesterday and I didn't want to put in a lot of work populating data. Also I'm pretty familiar with the companies so I felt pretty comfortable with removing companies. This was basically back of the envelope stuff. I could have put in a lot more work to do it properly and I would have arrived at the same conclusion.

 

Btw, the Dow data set is basically one that begs for adjustment. It's a small sample size with some big outliers: CVX and XOM on the income side and MSFT and AAPL on the debt side. S&P would have been much better and wouldn't need adjustment. But i chose to thank the OP for his work of putting these together instead of asking why he didn't do all 500 cos.

[Guest Cameron]

The logic is that nuance of how the debt comes to be. There's a world of difference between MSFT and AAPL and CVX and XOM. MSFT and AAPL are taking on debt to lower their taxes and have their debt covered 2x by cash in their bank accounts.

Then it would make sense to look at something like enterprise value which adds the debt to the market cap and subtracts out the cash or failing that at least subtract cash from the total debt. I am as suspicious as TwoCitiesCapital of selectively removing individual companies.

 

Of course working with net debt numbers makes a lot more sense. But I was pretty lazy yesterday and I didn't want to put in a lot of work populating data. Also I'm pretty familiar with the companies so I felt pretty comfortable with removing companies. This was basically back of the envelope stuff. I could have put in a lot more work to do it properly and I would have arrived at the same conclusion.

 

Btw, the Dow data set is basically one that begs for adjustment. It's a small sample size with some big outliers: CVX and XOM on the income side and MSFT and AAPL on the debt side. S&P would have been much better and wouldn't need adjustment. But i chose to thank the OP for his work of putting these together instead of asking why he didn't do all 500 cos.

 

I should have the S&P 500 data out sometime tonight.

[rukawa]

The logic is that nuance of how the debt comes to be. There's a world of difference between MSFT and AAPL and CVX and XOM. MSFT and AAPL are taking on debt to lower their taxes and have their debt covered 2x by cash in their bank accounts.

 

Then it would make sense to look at something like enterprise value which adds the debt to the market cap and subtracts out the cash or failing that at least subtract cash from the total debt. I am as suspicious as TwoCitiesCapital of selectively removing individual companies.

 

I'm thinking about maybe doing something similar with the S&P 500 but the shear amount of companies makes it seem unrealistic.

 

I could probably get this off of Bloomberg or failing that we could use this:

https://github.com/JeffFerguson/secqdb

 

I am going to try and see what I can grab off of secqdb...even though that is the harder route...because I've basically been itching to see how far the structured datasets can go. But if I can do it by the end of the week...Bloomberg it is.

 

As a result I am working on adding a net cash sheet to the document, I have updated the doc with historical NASDAQ returns, results are pretty interesting, I'm going to add as many exchanges with as much data I can get my hands on

 

If anyone has any other suggestions I'm game. I'm trying to see if there is anyway I can get full book value and market caps for every stock in an exchange to pin point cheap exchanges relative to each other as well.

 

Where are you getting your data from?

[Guest Cameron]

The logic is that nuance of how the debt comes to be. There's a world of difference between MSFT and AAPL and CVX and XOM. MSFT and AAPL are taking on debt to lower their taxes and have their debt covered 2x by cash in their bank accounts.

 

Then it would make sense to look at something like enterprise value which adds the debt to the market cap and subtracts out the cash or failing that at least subtract cash from the total debt. I am as suspicious as TwoCitiesCapital of selectively removing individual companies.

 

I'm thinking about maybe doing something similar with the S&P 500 but the shear amount of companies makes it seem unrealistic.

 

I could probably get this off of Bloomberg or failing that we could use this:

https://github.com/JeffFerguson/secqdb

 

I am going to try and see what I can grab off of secqdb...even though that is the harder route...because I've basically been itching to see how far the structured datasets can go. But if I can do it by the end of the week...Bloomberg it is.

 

As a result I am working on adding a net cash sheet to the document, I have updated the doc with historical NASDAQ returns, results are pretty interesting, I'm going to add as many exchanges with as much data I can get my hands on

 

If anyone has any other suggestions I'm game. I'm trying to see if there is anyway I can get full book value and market caps for every stock in an exchange to pin point cheap exchanges relative to each other as well.

 

Where are you getting your data from?

 

The companies websites. I'm sure there is probably an easier way but I don't have any programming experience thats why I chose the Dow first to see if this was feasible.

[rukawa]

The logic is that nuance of how the debt comes to be. There's a world of difference between MSFT and AAPL and CVX and XOM. MSFT and AAPL are taking on debt to lower their taxes and have their debt covered 2x by cash in their bank accounts.

 

Then it would make sense to look at something like enterprise value which adds the debt to the market cap and subtracts out the cash or failing that at least subtract cash from the total debt. I am as suspicious as TwoCitiesCapital of selectively removing individual companies.

 

I'm thinking about maybe doing something similar with the S&P 500 but the shear amount of companies makes it seem unrealistic.

 

I could probably get this off of Bloomberg or failing that we could use this:

https://github.com/JeffFerguson/secqdb

 

I am going to try and see what I can grab off of secqdb...even though that is the harder route...because I've basically been itching to see how far the structured datasets can go. But if I can do it by the end of the week...Bloomberg it is.

 

As a result I am working on adding a net cash sheet to the document, I have updated the doc with historical NASDAQ returns, results are pretty interesting, I'm going to add as many exchanges with as much data I can get my hands on

 

If anyone has any other suggestions I'm game. I'm trying to see if there is anyway I can get full book value and market caps for every stock in an exchange to pin point cheap exchanges relative to each other as well.

 

Where are you getting your data from?

 

The companies websites. I'm sure there is probably an easier way but I don't have any programming experience thats why I chose the Dow first to see if this was feasible.

 

Huh? You are going to go through the statements of 500 companies? Dude...I'll pull this from Bloomberg.

[Guest Cameron]

I have updated the file on my original post, you will now find the S&P earnings data, I wouldn't get your hopes up on more data from the S&P 500 from me at least until I find faster way to input the data, which is what I'll be researching tomorrow. Also it could be picked over to find companies with stable growing earnings.

 

I plan on updating the doc with more dow jones sheets, international exchange historical returns like i have done with the Dow and Nasdaq and S&P 500. I hope to do something with the NASDAQ 100 as well like i have done with the dow.

 

Again if anyone has any suggestions I'm all ears.

[Guest Cameron]

surprised this wasn't noticed but I didn't add DuPont on to the Dow list so the previous spread sheets are incorrect and will be updated shortly.

 

Soon I'll be adding a sheet that will allow you to change the stock prices to see how it affects the overall dow and if you took certain stocks out of the weighting what the p/e would look like and so forth. 

 

[Guest Cameron]

I added a new document that shows the us debt, interest payments, receipts, and outlays over the last 15 years. I was inspired to do it by the Ray Dalio book, can't wait for the second book. In my opinion Dalio uses the same framework as Buffett just on a Macro level. Countries are just like businesses but they extrapolate on revenue and expenses far more than SEC reporting.