Hacker News Reader

Comment by donaldguy | Hacker News Reader

I did a run at the (amateur) math off Google Finance and Crunchbase. With some companies I couldn't find's numbers outright omitted, and using old valuations, I totaled those corps filing as valued at $2.58 trillion in market cap., at least $131 billion of valuation and another $8 billion in investments in un-valuated companies. Assuming those investments hold value at least equal to the total worth of those companies in sum, we can reasonably assume the companies party to this brief represent value of at least 16% of US GDP.

https://docs.google.com/spreadsheets/d/1y0VtAf9OpjwU7Haf4dHY...

in case anyone else was wondering

EDIT: indeed it is > 15% even including only the publicly traded companies, for which the numbers of can be considered less speculative

View on HN

It's hard to compare market caps to GDP because GDP is kind of like "aggregate value of everything produced this year" while a market cap is "estimated total profits for a company from now until the end of time." So you might have a company like Walmart with $200b in market cap but ~$500b in annual revenue (3% of GDP by itself!) and $15b in net income. On the other hand there are companies like Adobe with $60b market caps and $6b in annual revenue and $1b in net income. So the GDP vs market cap comparison is somewhat apples to oranges, although there's no disputing the companies in the brief have a large impact on the US economy.

To be clear: it's not just hard, it is explicitly wrong.

A more correct measure is a much simpler one: the aggregate product of all these companies over the GDP. Note that this aggregate product is of course a component of the GDP, so you can express this neatly as a percentage. I see only one major accounting complication in constructing this metric, and that's the inclusion/exclusion of foreign products and subsidiaries. (See e.g. offshore revenues that are not repatriated, and therefore not part of the domestic product calculations -- I think?)

Another way is to estimate the net present value of America's GDP by multiplying it by 20 or so, which implies that it's roughly 1% of American wealth.

Alternatively, percentage of rents at office and industrial zoned land would be a good measure. That'd separate out national vs international activity, at the cost of being significantly more difficult to estimate and calculate. I actually don't know how well rents and imputed rents are tracked at aggregate levels and broken down by company, so this may very well be data that we simply don't have access to.

Out of curiosity, why ~20?

P/E ratios are around 20 on average I suppose. Meaning earnings are 5% of the market cap of a stock.

GDP doesn't directly correlate to earnings though.

Historically, $20 is the price of something that gives you $1/year. This, like, goes back to the medieval era.

It's just a rule of thumb as to what long run P/E ratios look like in the equity markets.

Are product numbers (domestic or multimarket) accessible somewhere to construct this? (Would revenue numbers suffice or is there more that should be factored in?)

Or is there an accepted measure of national economy that might play a sensible denominator for the sum I put together?

As I said, I don't know enough about economics to pretend to be making a reasonable comparison, I just went for the numbers I could find on a whim at 2am :-)

Thanks for actually knowing things

Could look at the GNI, gross national income. It's gdp plus net foreign income

Fair points, both you and akiselev; thanks for the clarification

I am not an especially economically-(or even really quantitatively) inclined person; I just see GDP bandied about as a metric of "size of economy" and market cap as "value of company". Revenue is probably locatable if someone wants to dig into quarterlies... I probably spent too much time on this exercise already

Here's a very, very rough ballpark estimate based on your data:

- $2.7T in market cap

- let's say average P/E ratio is 25 (P/E is approximately the ratio of company valuation to annual profits)

- let's say the profit margin of a typical tech company is 20%

- let's say 40% of a typical tech company's revenue is in the US

Estimated US revenue of the companies in your spreadsheet: $2.7T / 25 / 0.20 * 0.40 = $216b (about 1.5% of GDP, which is still huge)

By the way, I think your initial estimate of the total market cap of companies signing the brief is quite useful and informative, it just shouldn't be compared to GDP :)

It's actually pretty easy but for a first order estimate: http://wolframalpha.com/input/?i=total+2015+revenue+of+Googl...

I'm on mobile so I can't get the interface to work properly but it's much easier on desktop Mathematica or WolframAlpha. In the above figure, I only included names from the OP that I was certain were public after a quick glance and I didn't even get to Microsoft. The number is something like $500 billion a year but as a sibling comment to yours mentions, this includes international revenue that may never make it to the US.

Accurate data on this scale is hard to come by without a working Bloomberg terminal.

GDP is the total value of goods flowing through the economy (more or less), not the total value of all property so it's not a very meaningful comparison. By this measure, Vanguard's $1.7 trillion assets under management account for over 10% of US GDP but obviously that's incorrect, that money is just parked across a bunch of assets.

The US manufacturing industry is about $3 trillion a year in actual industrial production and the tech industry has way less leverage, let alone political clout.

Edit: the total market cap of the S&P 500 broke $15 trillion in 2013 and as of 2015, all of the companies in that index didn't even employ a fifth of US workers. Add in all the spending and investment property that doesn't make it into S&P figures (govt taxes + bonds, real estate, etc) and I bet the total number is probably >$100 trillion.

But that argument won't fly well: "Bunch of rich guys want to go against the will of the people"

Didn't Trump actually lose the popular vote? The USA isn't a democracy, it's a republic.

The "popular vote" is not the will of the people:

a) the rules of voting are clear, this is both the will of the people, and is also what the people are expected to accept as citizens. The time to denounce or change those rules is before your guy loses.

b) the popular vote is not the entire country, because not everyone voted. Not only do campaign strategies and other factors affect those numbers (to a degree that might make up the difference), but if you argue that people who didn't vote simply lost out, or don't count in deriving "the will of the people", then you could just as well argue that so did those who lost to the electoral college system.

When the union formed concessions were made... An Ohian's vote weighs, say 1.4, a californian 0.6. So depending on where you vote, your vote weighs more or less. I read Trump lost the popular vote by 3M. But then so did Bush.

I'm not American, my opinion is about as much worth as my passport.

When the union was formed, the big concession was the "three-fifths compromise": slaves counted for the purposes of vote apportioning, but weren't actually allowed to vote. Preservation of the racist institution of slavery was a driving force behind the unequal apportionment.

And California wasn't a state but a largely empty colony of Spain.

But didn't Hillary know it isn't the popular vote that gets her to presidency. The is no trophy for winning the popular vote! Why didn't she campaign in the Rust Belt states but went to California instead. It is really baffling...

> Why didn't she campaign in the Rust Belt states but went to California instead. It is really baffling...

Because, ironically, her campaign was convinced she had the election in the bag, and they didn't want to risk her losing the popular vote while winning the electoral college, because that could undermine her perceived "mandate".

She made a strategic mistake. But the "popular vote" keeps getting mentioned in the context which is validating her somehow. I see it more as another reminder of a big blunder she made not something positive to promote about her campaign.

still, those 97 companies do not represent a majority

Reply on news.ycombinator.com