Taxes and innovation: shout it from the rooftops

We find that taxes matter for innovation: higher personal and corporate income taxes negatively affect the quantity, quality, and location of inventive activity at the macro and micro levels. At the macro level, cross-state spillovers or business-stealing from one state to another are important, but do not account for all of the effect. Agglomeration effects from local innovation clusters tend to weaken responsiveness to taxation. Corporate inventors respond more strongly to taxes than their non-corporate counterparts.

That is from a new NBER paper by Ufuk Akcigit, John Grigsby, Tom Nicholas, and Stefanie Stantcheva, via Adam Ozimek.


Taxes don't matter: "Agglomeration effects from local innovation clusters tend to weaken responsiveness to taxation" - translation: in high-tax areas like California (compared to say low-tax Arizona), business like high tech will stay in Silicon Valley due to networking benefits. Paul Krugman is right. I've worked both in CA and AZ high-tech areas and it's like day and night: Arizona has great nightlife for college kids and pretty sunsets, but it's not a place for high-tech business, just ordinary business like catering to snowbirds and retirees and blue collar types who want to live there due to lower costs.

Bonus trivia: they water bluegrass in Phoenix. Most don't care about xeriscape.

PS--I know about Intel in Chandler, Motorola, etc etc etc. They don't count, they are islands onto themselves, and clean room work is blue collar work btw.

That's nonsense. Arizona has a vibrant start-up culture. American Express' North American technology hub is in North Phoenix. Amazon is opening a new campus there. What Arizona doesn't have is the cluster of "halo" companies like MS, FB & Amazon in Seattle, or Apple and Google in Silicon Valley. But if you are looking for investment money for your start-up, the venture capital scene in Arizona is very strong, and if you're an engineer worth half a grain of salt, you will receive job offers twice a week. Arizona is growing faster than most other states, and Phoenix is growing faster than most other metropolitan areas - primarily because so many businesses are fleeing high-tax states for Arizona's combination of highly-educated workforce for high wage jobs, coupled with the cheap labor illegal immigration allows for traditional low wage jobs.

Tech, like any of the high paying professions, is ultimately a tournament game.

If you’re in Arizona, you’re not playing.

Insert Obligatory investment banker in Hong Kong joke.

They sound so much like the early Paul Butterfield Blues Band and I love that hard core sound
so much. I can easily see why they walked away from the management.
If I had that sound I'd stick with it too. Maybe they should have set up in Chicago's South Side.
I need to find their records, if thay have
any. I play too, go type [ Maso's Blues ] into "Search" and see me.

To: "Just Saying"
Complete Nonsense on your part. Warehouses and call centers don't count. And, incidentally, looking venture capital for Arizona vs California comparing the 1999-2008 time frame to the 2009-2018 time frame, we see that Arizona went from seeing VC financing that was %1.5 of California's to %0.7 of that of California, a %53 percent reduction in percentage figures, compared to California, not exactly a ringing endorsement for a vibrant startup scene. Patents, BLS figures, and on an on, show California to be totally superior to Arizona in all things economic. At least Arizona has the Grand Canyon.

I caught that too, it seems to be saying something like:

"You may not notice taxes impacting innovation because clusters can distort the picture but taxes do matter too"

Or in other words if you have a cluster in one state with high taxes and another state with low taxes the low tax state will see more gain from innovation.

Is this Akcigit guy the new Acemoglu? His name sure stays in mind.

Bill Janeway argues that state support in various forms, including being a buyer in early markets, has been a crucial component of success of many tech companies. He recently had a long interview on Barry Ritholtz's Masters in Business podcast for those interested.

Generally speaking those advocating state support for research ignore the potential benefits if the money was left in the hands of the original tax payers, state money is assumed to be free.

But if you do have to have state funding of basic research, do it through prizes with the goal openly published for all to compete for. Research grants go to the people who reflect well on the research grant giver, not necessarily the people who can actually innovate.

Complete baseless, ideological nonsense on your part ChrisA.
Shameful delusion, really.

The exact opposite of that


So the state is better at picking winners than the market? Who can argue that the government pumping money into any enterprise doesn't give that firm an advantage? However, that does not prove that crony capitalism offers superior returns. Look at Russia. State investment in some firms have helped make the owners of those firms billionaires. For the overall economy, well have you been to Russia? Look at Japan. After some initial success, Japan lost much of its competitive advantage through mistakes in industrial policy. America's greatest hope is that China makes the same mistakes.

Re: "So the state is better at picking winners than the market? "

Right out of the gate, you offer the stupidest question ever uttered by humans.

I'll answer it. The state is literally INFINITELY better than private actors in choosing winners, given that we are talking about early stage, capital intensive, knowledge development investment of a long time horizon, which the private sector plays absolutely no part, and which the state actors are the ONLY players. The state, in essence, is the investor of "first resort." So, 1/0. You are infinitely wrong, I am infinitely correct. Everything else, past your first sentence is absolutely irrelevant--in all possible universes.

Good, LORD, you are the stupidest person in human history.
You know nothing, you fool of all fools.

Oh. It’s parody.

In a saner world, DanC would be a parody.
Anyhoo, he might, and that's awright.

What are you doing on an economics blog if you know nothing? And what you think you know is wrong?

When you get out of the 7th grade, please make more informed comments.

"Knock knock"

"Who's there?"

"The Stone Age"

"The Stone Age, whaaaaa?"

"Yeah, that's right, the Stone Age. We just want to bring DanC back to show him this thing we discovered: fire. Ya can heat your meat with it."


So, there ya go, DanC. Go heat your meat.

Bill Janeway interview, as well as recent interview on recent Freakonomics of Mariana Mazzucato, are simply attempts to reach the pinnacle of Captain Obviousness.

The devil made me do it! That's the explanation when companies such as Apple, by shifting production overseas and the creative use of accounting rules to shift income to tax havens, avoid U.S. taxation. That's the explanation when firms threaten to shift operations to low tax states in order to receive tax concessions from the states and avoid state taxation. As Ray points out, the most creative place in the U.S. is Silicon Valley, which, last I heard, is still in California, a very high tax state. Of course, devolution, shifting government responsibilities from the federal to the state level, just encourages states to engage in predatory practices to induce firms from moving from one state to another.

"just encourages states to engage in predatory practices to induce firms from moving from one state to another."

Yes! Competition is good.

“higher personal and corporate income taxes negatively affect the quantity, quality, and location of inventive activity at the macro and micro levels.”
I sure know some redistribution profiteers that will find a hundred ways to rebut those findings. The authors of this paper are sure entering into a lion’s den. But, good for them (and for us) they dare try.

Anf New York and Boston also appear to still be ahead of Phoenix and Austin last time I checked, in high tax locations. I have not read the paper so maybe they get their efffect after taking account of all sorts of other things, especially agglomeration. But the quoted statement about the apparently overwhelming impact of taxation seems way overdone. Another obvious point is that some of these higher tax states also have well-funded high quality universities, although in the cases of New York and Boston the more important ones are private, and in Silicon Valley, Stanford is crucial, although Berkeley and UC-Santa Cruz have played important roles (and UT-Austin pretty good although possessing a large endowment funded by Texas oil operators).

New York is finance, fashion and media. There’s a large signaling aspect to the sales side of all three. This has overridden any large fallout from taxes. You’re not going to locate an investment bank in Milwaukee. There’s also a path dependency operating here from network effects. Note, however, that hedge funds have begun to move outside the city limits, due to taxes.

Silicon Valley is mostly driven by California refusing to honor “Non compete” clauses. This does not get mentioned nearly enough. This was critical in the differential in the rate of growth in tech sector NE vs California. Rest is a combination of amenities, weather, and network effects.

I don’t know what industry Boston has vs other cities. Irish crime? There’s medical device manufacturing west of Boston. There’s Wayfair. BCG.
Unless federal tax payer funded/subsidized higher education is an industry (Harvard, MIT, Tuft, Boston College, Boston U, Northeastern....).

Well, they all have medical schools and hospitals...Boston is not only device manufacturing.

Insurance and financial services.

Thanks, you made me look up the numbers and it’s more interesting than I thought it would be.

Boston population : 673,000
Students in higher education : 152,000
Employees in higher education: 69,000

Total : 32.8% of the city either directly employed in higher ed or students. Probably less due to overlap / not all students counting as “population.”

What happens to Boston If higher ed subsidies go away?

Re: What happens to Boston If higher ed subsidies go away?

Subsidies will never go away for this. NEVER! It's just a matter of what kind of subsidies.

Boston has several education intensive industries like finance, high teck, medical and not to forget education itself.

Boston is very good at creating high paying jobs, but poor at creating run of the mill middle income jobs. Doctors earn less in Mass than any other state, but stay there becauseof the quality of life.

Overall.Boston has an inflow of highly educated individuals while the middle class are moving out because they can not afford the city. It
is like my sons sister-in-law. She left a job as a hair dresser in Mass
to be a dealer in Los Vegas and average income in both states rose.

By the way, Massachusetts is no longer taxachusetts.

It is right in the middle of the states on its tax burden.

You need to look at the Boston MSA. Just looking at Boston, your "education" numbers are missing Harvard and MIT.

As if Boston's higher education industry is primarily concerned with supplying higher education to people from Boston.

Detroit wasn’t primarily concerned with selling cars to people in Detroit either.

As always, you miss the point entirely.

Take out the student population and you get 13% of Boston's population is employed providing higher education. From I'm getting Boston's GDP is about $400B.

So 152,000 students, Google says list tuition + room/board/books is $70K per year. That would be $11B which should be paying the salaries of that 13% of people....that's being generous because most students won't pay list price but get some scholarship plus some students will be part time.

What's kind of curious about this is $11B is only about 3% of Boston's GDP. So the bad news is if higher education vanished tomorrow (say free brain downloads eliminates the need to have colleges), 13% of Boston goes unemployed. Good news Boston's GDP only falls by 3% and perhaps will increase by even more since it seems the higher ed labor force is not very productive. If that could be released into the rest of the economy Boston would be even bigger.

The hilarious part is where you think tuition is equal to university budgets.

Jesus Christ.

Havards annual budget alone is $4.5 billion.

MIT and Harvard alone are over $8 billion and it doesn’t include most grant funding/federal funding. Which pays for the graduate research.

Try again.

Grant funding would still happen as that's not so much higher education as R&D. In a world where higher ed. could be replaced with brain beaming we would still have to do research to discover new knowledge. College budgets are a bit deceptive because they would also include things like investments for new buildings etc. I figure the tuition at least must cover the actual physical staff of the college plus the operating costs of the college.

Keep in mind Boston's college sit on some of the most valuable real estate in the world. If tomorrow Harvard noticed it didn't need dorms or classrooms anymore, it could turn around and make a fortune changing the campus into condos, or corporate office buildings etc. Another reason I suspect a pull back in higher ed would have less inpact on Boston than one might think

Silicon Valley is mostly driven by California refusing to honor “Non compete” clauses.

Really? All North Dakota has to do to become Silicon Valley is get it's judicial system to adopt CA's view of non-compete clauses? I'm highly skeptical of this claim anyway. I'm sure Lyft is helped somewhat by people leaving Uber and going there but would there really be no Lyft otherwise?

There’s a large signaling aspect to the sales side of all three. This has overridden any large fallout from taxes. You’re not going to locate an investment bank in Milwaukee.

This seems exceptionally weak. Investors won't trust someone with their money if they are coming from the midwest where taxes are lower, unless he is Warren Buffet? Therefore taxes are high in NYC because this is how Wall Street signals to customers to trust them with their money? Yet how about the carried interest rule and hedge funds, do you think they lobby for or against taxes?

I do think you're right about path dependencies and network effects overriding tax policy. What I think the paper is saying, though, is absent that tax policy matters. If NYC has high taxes and so does London and both are financial clusters, but NYC has higher taxes, then London will enjoy an advantage.

Did you skip the paragraph ?

New England lost out to Silicon Valley in large part due to the cumulative effects of non compete clause enforcement. This is not controversial. You’re torching a straw man. Also Uber to Lyft comparison ignores everything that makes it “successful” that does not literally have to do with ride hailing. Where do you think they got their engineers ?

The rest of your comment makes me think you don’t understand what an investment bank is.

The customers are not Joe Sixpack, it’s Company A looking to buy out Company B or preparing for an IPO.

Carry loophole is a federal issue.

The point is that nyc as a finance center is not as susceptible to tax rates as a factory would be on the border between Illinois and Indiana. Part of this is the signaling aspect, absolutely. For IBankers, not Warren Buffet.

It probably depends not just on taxes but on a state’s overall attractiveness to innovative workers and businesses. For instance, California can get away with high taxes because it is very attractive to the most innovative workers due to the great weather and cosmopolitanism. Other places without those resources might be able to get away with higher taxes if those taxes are invested in quality of life amenities and do not disproportionately target innovators. But your average state or local government that’s charging high taxes to fund giant pensions for retired government workers is going to suffer.

But your average state or local government that’s charging high taxes to fund giant pensions for retired government workers is going to suffer. As here.

All things being equal, less taxes is good

But is it good practice to shout from the rooftops "less taxes" in a one trillion dollar deficit year?

I mean, I thought Tyler Cohen was a little better grounded than Grover Norquist.

Only method ii) has anything to do with federal taxes.

Let me guess, you didn’t read the article.

I briefly considered whether that was important, and I decided it was not.

No one here experiences state taxes in isolation, we all experience "the stack" from local to federal.

Thus, we all experience deficit spending, and the idea of this particular strata of taxes should be lower is .. again unrooted.

$1T deficit? Looking at the Monthly Treasury Statement through July, revenues are slightly ahead of last year, while expenditures are about $145B higher. We might end up around $784B versus $683B last year - and we were told that the Tax Reform Act of 2017 would cost about $190B per year. Maybe tax cuts do pay for themselves with higher growth.

And of course it would be hard to name an "innovation cluster" that was founded without some sort of federal funding.

So...we should levy federal taxes on people in the Midwest to raise real estate prices in Palo Alto, New York, and Boston?

Median home price in Palo Alto is over $3.3 million dollars already. Median.

This definitely funnels money into my pockets, but as a policy choice it makes little sense.

It's interesting that you should ask that, as if it is a rhetorical question.

Silicon Valley has arguably been the center of technological innovation of the century, but what are its roots?

The electronic demands of military industrial complex.

A cluster focused in on cricket-breeding? Well, maybe some USDA money set things in motion. Anyway, you can hear those crickets, fo sho!

I agree but is this a chicken/egg issue here? There's all types of Federal funding everywhere. If one day some place in North Dakota erupts in economic prosperity and rapid growth, I'm sure you'd be able to find some Federal funding that place enjoyed.

That doesn't quite answer the question of why did an 'innovation cluster' erupt in CA but not ND? Workers in cubicles punching out code for military industrial complex contracts presumably could work from either location. And it hardly explains other clusters. Did NYC's fashion cluster get its start by giving the federal gov't designs for military uniforms?

Maybe my impression is wrong, but I think people figured this out ~20 years ago and started trying more consciously to spread research dollars around the country.

I think that's fine.

I think State University Systems should also take care to expand into less-developed areas.

I don't think you can avoid clusters. You can spread the 'seeds' of clusters around as evenly as possible but after time you'll find industries/innovations clustering.

We would have a much better politics if the right and left negotiated on how much spending we do, and this year's spending is next year's tax. It's the ultimate PayGo.

I guess you have to say explicit stimulus exempted.

But the bottom line really is that treating tax as an independent variable has become the root of our dysfunction.

I read through the first few pages and the summary of findings and didn't find anything about R&D tax credits being accounted for. In my state, there is a tax credit for activities which fall into "R&D". This feels like its a major oversight on the researchers part, unless I just didn't read far enough...?

Minor quibble.

Edit for accuracy:

Change from: "Agglomeration effects from local innovation clusters tend to weaken responsiveness to taxation." to "Agglomeration effects from local innovation clusters tend to entirely eliminate responsiveness to taxation." OR "Agglomeration effects from local innovation clusters tend to entirely overwhelm and bury the responsiveness to taxation."

Cleveland once upon a time had more millionaires than any location in America. From the middle 1800' to early 1900's Euclid Ave in Cleveland was wealthier than any section of New York. Some of the homes were over 50,000 square feet with 100+ servants and staff. John D Rockefeller was the most famous and richest. Most of those homes are gone now. Cleveland is now one of the poorest cities in the country.

Detroit was the sixth richest metro region in 1970 now it is below fifty.

San Francisco/ San Jose is about the 7th largest metro area in the USA measured by GDP. They are having their day in the sun.

Compared to many other areas, silicone valley is less diversified. What happens if they lose their advantage to China, Korea or India the way Detroit lost their edge to Japan and increasingly Korea?

What if high taxes, government regulation, etc cut into margins. How will employees react when some of those generous perks go away with the advent of those narrowing profits. Employees in the auto industry didn't react well.

Higher taxes and greater regulation mean narrower profit margins. That it opens the door for competitors to enter the field. Ireland, India, China, Korea, etc all have very bright people who are willing to step in. And they won't face the ever increasing costs of living near the bay.

What is the MC of the intellectual output of the tech firms? Today the model is that they come up with an idea where the startup cost is high but the MC is very low. Almost like printing money. You charge a customer $300 for something with a MC near zero. What happens when competition, taxes, and rising regulatory costs pushed MC close to MR? Ask the people in Detroit, Cleveland, etc

The idea that people in CA don't care about taxes is simply wrong and should be absurd on its face. I know tons of people who leave CA for other places. Bay area has a high average income, but nowhere near NoVA and the prices are three times as high in the Bay area. If you can't find a job in NoVA you're not trying.

Nonsense that SF Bay Area/Silicon Valley is less diversified in its mix of tech sectors. It's breadth, as well as depth, of tech sector activity is unrivaled, and is its primary strength.

"For years Silicon Valley dismissed Chinese tech firms—first as an irrelevance, then as industrial spies and copycats. Most recently China has been seen as a tech Galapagos, where unique species thrive than would never spread abroad. But as our Schumpeter column explains this week, China’s technology industry has been catching up far faster than expected. American venture capitalists return from trips to China blown away by its energy. China’s government has set a goal of becoming dominant in artificial intelligence by 2030."

Wow. You've entirely wiped out, and totally, contradicted you earlier comment (DanC: "So the state is better at picking winners than the market?") to me earlier.

Thank you. You destroyed your own ideological claptrap.

Look Moron, I thought you could read.

My earlier comment was that "After some initial success, Japan lost much of its competitive advantage through mistakes in industrial policy. America's greatest hope is that China makes the same mistakes."

Technology transfer is much easier than development. Second much of the Chinese government is acting as private investors - government officials are often seeking private gains. Think of it as laundering tax revenues through a company into the pockets of bureaucrats. Is it the best investment for the general population? Probably not.

The Chinese government can pick some industries to concentrate on. There is zero, zero, zero reasons to think that those government officials have superior skills at selecting which technologies will, in the long run, be successful.

However many firms in China are competing fiercely and winning in the marketplace. Not every Chinese firm is funded by the government.

Now ask your mommy for a glass of milk and a healthy snack to go with your overdue meds.

Again, you’re the idiot, conflating the aggressive Chinese support of domestic tech service alternatives with the USA approach (also involving gov support) to developing new system architectures to create and support future industries, as well as revolutionizing those already established.
So, in this regard, the Chinese approach is a total disastuh [sic].

Oh, wait...Tencent, Alibaba, Baidu, and others. Maybe not such a disaster.

The difference in US and China approach? The Chinese like to stay in office. The American model is you wait to leave government service than cash in your favors.

Now take your meds and take the short bus tomorrow.

Nice. I see that you pulled that work style generalization out of your keister.

Anyway, gonna go enjoy a lovely IPA. Toodaloo!

Fun fact: Rockefeller left Cleveland when they created a tax bracket that only he fell into.

Time to annoy people once again with my tax proposal.

1. Eliminate taxes, the Fed will create a set amount of money each year and give it to the Treasury. After that buying and selling bonds will be based on their usual interest rate policy to keep inflation & unemployment low.

2. All Federal spending is deficit financed.

3. Instead of fines and penalties payable in cash, they will take the form of mandated long term bond purchases (example, Wells Fargo might have to buy $75B in 50 year bonds and hold them for at least 35 years). Instead of social security/unemployment taxes workers will have to buy bonds which they may, but don't have to, sell at retirement age.

4. NEW - IMF style 'Super-Senate' which can override any decision the Senate makes. Each state's representation will be proportional to their holdings of bonds. In general small states will have less power, like in the House, but this will also be contingent on how willing the state is to tax itself to buy bonds giving it more say.

While you still have whatever dead weight costs of state and local taxes, you'd immediately benefit by eliminating all incentive issues from the much larger taxes at the federal level.

Filling in for Ben Cole today?

Someone has to do it. (Have to say I don't get the reference...)

Ben Cole also seem to subscribe to MMT. He is using Japan as evidence it works.

In my experience, Tyler uses "shout it from the rooftops" for something that is tough to understand but undeniably true. Is that the case here? I think that taxes do discourage R&D, but I'm not sure how I would prove it. Research in this area is going to be very fuzzy, like minimum wage research.

My take is that since we have innovation clusters in areas that are known to be tax heavy while tax light areas seem to do poorly as innovation clusters, one might jump to the conclusion that taxes have no effect on innovation clusters or even that higher taxed areas invite innovation clusters. It's easy to come up with an economic theory that might back such an idea up....for example if you have a healthy safety net a worker might be more open to taking a risk in creating a startup knowing even if the whole thing fails his kid is still going to get treated for cancer.

The study he is citing seems to be saying, in an awkward manner, that this isn't quite the case. You can get innovation clusters in high tax areas but higher taxes do harm them hence you should still be trying to keep taxes as low as possible.

The weakness in the argument "Lower taxes= Higher innovation" is that it treats a large outcome like "innovation" as the result of a single variable.

But of course, California did not become a leader in innovative technology due to any single magic bullet.

Innovation results from a cumulative effect of education levels, culture, infrastructure , which are actually fueled by taxes.

Lower taxes always translates into a sort of free lunch proposal, whereby all manner of sweet and wonderful things will come about in a cost free and pain free manner.

Just suppose a state happened to have a boatload of money for some reason and established all those things you talked, infrastructure etc. and enjoyed an innovation cluster forming.

In contrast California had to do it the hard way by actually building those things without any gifts from the sky.

The argument here would be the lucky state would have an edge with its innovation cluster.

Now the reality is, I think, taxes follow the cluster. CA has enjoyed multiple 'innovation clusters' in its history. Gold, agriculture, tourism, Hollywood, aerospace, tech, etc. etc. The success of these clusters created demand for things like education, infrastructure, etc. which brought taxes which then bring more clusters.

Geography helps (and hinders), too.
California has a nice high desert, which is perfect for aerospace testing (yielding an aerospace cluster).
It also has a wonderful Mediterranean climate, which draws expertise from around the world to live in its agreeable, salutary weather, also a climate that leads to high air pollution (dry season high pressure coupled with topography), which, in turn, leads to higher regulation and business costs.

"Agglomeration effects from local innovation clusters tend to weaken responsiveness to taxation."

Common sense (see, e.g. San Francisco, Silicon Valley, Los Angeles, New York City) argues that this factor swamps tax in magnitude. People aren't inventing like crazy in low tax states in the U.S. Southeast, or in tax havens.

Yup, and the “tend to weaken” qualification yields an extraordinary understatement.

How does this post relate to the “extractive institutions” post from 2 days ago? (As an underwater basketweaving expert, this is all way above my pay grade).

Bloomberg’s most innovative list from several years back showed Right-To-Work states dominating the bottom of the list. Not one R-T-W state made the top ten.

Right-To-Work or not Right-To-Work is not the variable determining success or no success. It's a correlate to other things, though. We leave that to the wonderful readers of MR to uncover.

In 1960, the U.S. had a five-year life expectancy head start on Cuba. Half of the states that became R-T-W in the forties and fifties have since fallen behind Cuba’s life expectancy by an average of two years.

Fair enough, or maybe not.

Comments for this post are closed