Counter fact: In our city district, which is the richest and biggest district of our 600k developed city, we decided to turn off the street lights at night on purpose to help with sleeping better. There is no street criminality, people feel safe without street lights. Our city is the richest in our country, which is at the top 5 in the world
It feels like it is something that could show broad long term accuracy over say a decade but short term the noise level would make it difficult to extract a signal.
Even then I would guess there would be a lot of other leading signals than lighting that would also correlate.
I think in a highly financialised free market like the US the executive is pretty powerless to really hide important economic data like this, even within their own country.
There would be so much alpha in knowing stuff like the true employment rate that private agencies will be extremely well-funded to collect this data (in fact this probably already happens for some data, there might be alpha in having a second opinion even if you think the government data is trustworthy).
I think the worst-case outcome would be that the mass populace doesn't have access to the info because it's paywalled. But to the extent that journalism continues to exist, journalists will know the employment rate, GDP, that kinda stuff.
Light at night would overstate success for the incumbent US population actually. Since (contrary to popular belief) we're significantly less nationalistic than these other countries and have a much larger and more successful migrant population.
I spent a lot of time living in China. Nobody believes the government figures. But I'm also skeptical that using artificial light as a proxy for economic growth is rational, particularly when you realise that Chinese people overwhelmingly live in vertical high density buildings and the amount of light used when moving from last-gen 'heavy industry' to next-gen 'value add'/'light industry'/'design work'/whatever is going to be reduced.
Therefore although I am a big fan of the Economist and like the idea, I think the premise of this particular study may be somewhat flawed.
Where the article states "the mismatch between satellite and GDP data did not appear in dictatorships until they were too rich to receive some types of aid" I think what they may be discovering is "when people move in to dense modern housing and shift to white collar work the model breaks down". There are other factors too: more modern lighting is more efficient, people increasingly socialize through phones, and outdoor living spaces are reduced in relatively inhospitable climates, somewhat limiting light pollution.
Thinking back to first principles, the majority of outdoor light pollution is probably from freeways and city centers, and if you proxy that with economic growth it's probably significant as a pre-emption at a certain phase of transition from agricultural/low-development-level economy through highly developed economy, but becomes irrelevant rapidly once those development prerequisites have been achieved.
It doesn't help that this guy is trying to sell a book.
> the amount of light used when moving from last-gen 'heavy industry' to next-gen 'value add'/'light industry'/'design work'/whatever is going to be reduced
Not to mention the automation of heavy industry leading to "dark factories": some Chinese factories are so completely automated now that they don't bother turning on the lights in large chunks of them. See https://www.youtube.com/watch?v=MCBdcNA_FsI
>using artificial light as a proxy for economic growth is [irrational]
it's interesting you pick on this detail. I'm of a mind that "not free" govts control information so carefully and lie about their statistics so thoroughly that we can use that discrepancy to establish proper weights for our measured lighting scale.
I remember reading about how regional Chinese governments measure economic growth, given that the numbers they have are doctored. One proxy was electricity usage.
Now obsolete Le KeQiang Index (LKI): power, rail freight volume, bank loans which were harder to manipulate. But this was like pre 2005s when Li was in charge of Liaoning before became premiere. Western analysists picked up on it form wikileaks and started using it as proxy - relevancy dropped when heavy industry era ended in mid 2010s. It's 2025, PRC heavily digitalized in last 20 years, frankly it's pretty absurd to believe central gov has significant problems with local gov metric opacity these days. As in local govs will still try to game, but central gov / NBS or relevant stats bureau has been adjusting down accordingly because they have access to stupid amount of proxy data now.
That's actually mentioned in the article. The information came from a diplomatic cable. It's very damning, although as evidence it does come down to one (important) person's view.
"I'm also skeptical that using artificial light as a proxy for economic growth is rational"
I found myself wondering if it was a lagging indicator. Hopefully the peer review process would have flagged these issues if they were serious. I didn't see the venue mentioned though.
It would be pretty easy to validate the model, I think: take Eastern Europe, South Korea, Norway, Ireland as examples of countries where the economic growth since 1980 was very obvious, and most of it corresponded to a democratic society. Then take the US, Japan, Germany, France, UK, Sweden as a control group, which was already pretty developed by 1980, and check their trends in light pollution vs GDP, or whatever.
(1980 is an arbitrary date, but before the fall of the USSR and thus the explosive growth of the Eastern Europe, and when shots from orbit likely became easy to obtain.)
Your rationale doesn't explain Japan, which, due to scarcity of land, has some of the most vertically dense cities in the world and yet its major cities are some of the brightest.
Japanese cities aren't particularly tall, especially when compared to Chinese cities. Even the busiest parts of Tokyo often have buildings that are just a few floors tall and single family homes are everywhere throughout the city.
And once you're in other big cities, this becomes even more true. It's common to see single floor businesses and buildings right in the busiest parts of town.
Yeah, I bet it isn’t a simple linear model, at least. But I also wonder if a model that takes the effects you’ve identified into account could be trained. I guess we’d have to have some historical source of the true GDP numbers, though.
The premise of the study is that light is a flawed but easily obtainable metric that correlates with GDP growth. There are no doubt lots of other metrics that go into estimating economic growth when self-reported numbers can't be trusted. But those take money and expertise to collect, and are probably mainly available to intelligence agencies.
I agree with your skepticism of the method and it's good to explicitly list these things. But I think the authors of the research would also probably also agree that the method is far from perfect.
> when people move in to dense modern housing and shift to white collar work the model breaks down....more modern lighting is more efficient
These should apply equally to dictatorships and democracies right? Or at least it shouldn't correlate with the dishonesty of the regime so the model can factor it out.
> people increasingly socialize through phones
You still need light for most forms of economic development. I've been to a few places where it's almost completely dark at night and people communicate on phones. But the economic centers, for example where people congregate for night life, have lights on.
Cities like Tokyo, Seoul, Singapore, and New York all have high-density living, yet they do not exhibit the same divergence between satellite-observed light and reported GDP. If urban density were the primary cause of the mismatch, it would appear across both democratic and authoritarian countries.
Similarly, gains in energy efficiency, such as widespread LED adoption, are global and not limited to any regime type. The same applies to economic transitions from heavy industry to services and behavioral shifts toward indoor or screen-based activity; these are common across modern economies. However, the study finds that the light/GDP mismatch emerges selectively in authoritarian regimes once they pass the income threshold for certain types of foreign aid.
This pattern suggests that the divergence is not driven by modernization effects alone, but rather by systematic incentives to inflate economic data.
Yes the “good hitler years” were a lie and so are all the “effective dictators”
And the fact that no one just assumes that is weird. In general, let’s imagine you had a politician who took power of a country that was recovering, and then by the time they left power their country was a literal pile of rubble and they shot themselves and their family in the head in order to avoid the consequences of their own actions… you’d assume that any positive story about them is probably bullshit. But for some reason the moment it’s Hitler everyone’s got an excuse.
And if someone accidentally killed 6 million of their own citizens we’d naturally all recognize them as one of the worst politicians in human history, but for some reason when they kill 6 million of their own citizens on purpose it’s not a raucous failure that deserves endless ridicule.
Lee Kuan Yew (whom the late polymath and Berkshire vice chairman Charlie Munger greatly admired) and Park Chung Hee are two examples that quickly come to mind. I distrust technocrats and dislike dictators, but pretending every dictatorship has been a disaster for its people is short-sighted.
Also: originally, “dictator" was a magistrate appointed to hold sole power for a limited time during emergencies. This original, Roman sense of the word did not carry the negative connotations it has today.
Search for "90s in Russia." [1] 20+ years under Putin's control brought it from that back to an undisputed superpower. Benevolent and effective dictatorship is probably the most effective form of government in terms of producing results. The problem is that when you end up with self centered, incompetent, or malicious dictators (all which somehow often go together at all at once) it's also the most effective form of government in terms of collapsing countries.
The same is true all the way back to the Ancient Empires which were also usually ruled by dictators. The era of Marcus Aurelius was an absolute Golden Age in Rome. Yet the era of his son all but ensured the collapse of Rome. Of course the same is becoming increasingly true of democracies where political messaging has become effective enough to regularly make people vote in highly irrational ways.
"Undisputed superpower"? That seems like a stretch for a description of modern Russia. Outside of having a stockpile of nukes, does anyone actually think that modern Russia is a superpower, with a failed space program, a flailing military, and rather questionable economy? It's not exactly a bastion of political power, either, beyond a rallying point for "nations who hate the US/west" or "nations who can take advantage of its power economic position to their own advantage and don't care about US opinions" (India), neither of which are really an indication of innate power for Russia.
Honestly, regression to the mean is a stronger explanation here than "benevolent and effective dictatorship".
It is ridiculous to state that Putin made Russia great again. Russia, by all economic metrics, is moribund. By social metrics, the birth rate is one of the worst in the world and they don't even have the privilege of dealing with the problems of massive immigration because no one wants to move there. By military metrics, it can't even fucking beat its weaker neighbour without devolving to outdated meat-grinder tactics with drones sprinkled on top. Putin does that because he's collectively punishing the military for failing the initial invasion. Russia can't even have a functioning civil society because everyone is too scared to do anything for fear of upsetting the regime.
I'm flabbergasted that you look at 2025 Russia and consider the word undisputed apt. How ... narrow-minded.
Which economic metrics do you mean? Here [1] is a large series of economic metrics on Russia. I linked to real wage growth because it's one of the most important, and it's been sharply increasing for decades, like most economic metrics. I completely agree on the fertility rate issue. That will be their primary challenge over the coming decades, though I am curious what it would be if we exclude the 'missed decades' generations from the 90s and early 00s. A quick search turned up little. In any case this will also be the main challenge for most of the developed world over the coming decades.
And I think Ukraine is obviously going to be a major turning point in history. Ukraine created an absolutely massive army by combining massive scale forced conscription alongside preventing men of "fighting age" (18-60) from leaving the country, and they're similarly being armed with hundreds of billions of dollars in Western arms - far more than we even supported the USSR with during WW2. And yet Russia, a country that could barely hold itself together in the 90s, and is under severe sanctions, is winning. We're looking at the absolute end of any concept of a unipolar world, and I think that's a great thing for everybody.
Not that Hitler is an example of one but there is actually a long history of good dictators. Remember the original dictatorships in Rome were time boxed (among other things) to overcome crises. England had a similar idea in the form of its protectorates. I don't like everything Cromwell did for example but he absolutely was a dictator in all but name.
How could they possibly lie? Don’t people just report the facts and then the facts show a bad jobs report and the labor economists who used to write the jobs report gets fired and replaced by a lying stooge and. Wait.
I cant imagine that light is a good proxy for growth, unless there were very good baseline maps that have been properly calibrated for the significant changes in the types of lights used and how they are bieng used......"dictators" lying
should be a simple presumption.......
which historicaly was confirmed by basic spycraft, in that least sexy of industrial chemicals ,hydrochloric acid, is still a very good indicator of total industrial capacity as it is used for all primary industrial production, and is the most used chemical world wide, but is now bulk shipped, instead of bieng produced localy.
Back to light as a proxy and an indicator, the flip side would be to restrict light, and hide industrial locations, as is common in war zones....so....back to spycraft 2025
as to China....they realy REALY are building out there electrical grid , road and rail networks, and no one can doubt that they are not producing massive amounts of everything, andwhile they have an incetive to exagerate there growth, there wester adverseraries have an incentive to lie about the same thing.....
the basic truth is that the world is definitly heading towards manufacturing overcapacity for everything, but that this is perversly bieng treated as a bad thing.....
The frequency of Martinez Democracy vs Autocracy study reposts, especially in western MSM, borders the PRC collapse retardation news cycle.
Nvm it got dismantled by various people who use nightlight metrics for living. NBER also did a sophisticated night lights study of PRC in 2017 - same year as OG Martinez study using naive methodology as Henderson original work from early 2010s - and found PRC's reported GDP was actually _underestimated_ using night light data. But useful idiots will gobble up authoritarian gov inflate narrative.
For obvious reasons this study by multiple staff economists at the FED doesn't get any attention in western MSM, but one by an assistant prof of public policy somehow does. Also note scope, multiple staff economics focusing on PRC, versus assist prof who managed to collate data for every country. Almost as if it's good self marketing considering Martinez keeps updating study with same rudimentary methodology every few years to get clicks.
E: I'm guessing this is now going to make rounds with the Shih/Elkobi's 1/3 of PRC local gov spending all their revenue on debt repayment wank graph being popularized recently that's comparably retarded.
What's there to elaborate. First version of Martinez study (when researcher was an assistant prof in public policy) has methodology criticized to death way back in 2017 by experts who specialize in satellite imagery, remote sensing etc. The same year NBER, National Bureau of Economic Research aka... one of the preeminent eco analysis think tanks US gov draws from had 3 credentialled staff economist ran a more sophisticated light study and alleged PRC (China) underreports economic activity.
Yet every year since 2017 the Martinez study gets play in mainstream media (MSM) because autocracy incompetent vs democracy plays well. In 2022 Martinez updated the study (I think also 2019, 2024 i.e. milked forever), with same methodology issue said insisted trend continues. Cycle repeats usually synced to some other retarded reporting to PRC economy going to explode (will also be milked forever).
In this case I'm assuming the chart making it's way around reddit and now socials on Shih/Elkobi study that inflate PRC subnational debt situation by conflating principle+interest (tldr bad methodology using non-standard metric to exaggerate debt risk), which like Martinez has been called out back in 2022 after publication. Customary in these discussions of unserious providence is bringing up the unserious Martinez study, which useful idiots continue to take seriously.
https://www.cbsnews.com/news/trump-fires-commissioner-of-lab...
Even then I would guess there would be a lot of other leading signals than lighting that would also correlate.
USA (2013-2023 CAGR: 2.3%) 2014: 6.2% 2015: -5.3% 2016: -1.8% 2017: 15.2% 2018: -4.9% 2019: 4.5% 2020: -5.4% 2021: 6.7% 2022: 14.5% 2023: -3.6%
China (2013-2023 CAGR 7.9%) 2014: -1.7% 2015: -1.2% 2016: -5.1% 2017: 53.3% 2018: -1.0% 2019: 7.5% 2020: 6.5% 2021: 11.4% 2022: 4.2% 2023: 10.8%
There would be so much alpha in knowing stuff like the true employment rate that private agencies will be extremely well-funded to collect this data (in fact this probably already happens for some data, there might be alpha in having a second opinion even if you think the government data is trustworthy).
I think the worst-case outcome would be that the mass populace doesn't have access to the info because it's paywalled. But to the extent that journalism continues to exist, journalists will know the employment rate, GDP, that kinda stuff.
I spent a lot of time living in China. Nobody believes the government figures. But I'm also skeptical that using artificial light as a proxy for economic growth is rational, particularly when you realise that Chinese people overwhelmingly live in vertical high density buildings and the amount of light used when moving from last-gen 'heavy industry' to next-gen 'value add'/'light industry'/'design work'/whatever is going to be reduced.
Therefore although I am a big fan of the Economist and like the idea, I think the premise of this particular study may be somewhat flawed.
Where the article states "the mismatch between satellite and GDP data did not appear in dictatorships until they were too rich to receive some types of aid" I think what they may be discovering is "when people move in to dense modern housing and shift to white collar work the model breaks down". There are other factors too: more modern lighting is more efficient, people increasingly socialize through phones, and outdoor living spaces are reduced in relatively inhospitable climates, somewhat limiting light pollution.
Thinking back to first principles, the majority of outdoor light pollution is probably from freeways and city centers, and if you proxy that with economic growth it's probably significant as a pre-emption at a certain phase of transition from agricultural/low-development-level economy through highly developed economy, but becomes irrelevant rapidly once those development prerequisites have been achieved.
It doesn't help that this guy is trying to sell a book.
Not to mention the automation of heavy industry leading to "dark factories": some Chinese factories are so completely automated now that they don't bother turning on the lights in large chunks of them. See https://www.youtube.com/watch?v=MCBdcNA_FsI
it's interesting you pick on this detail. I'm of a mind that "not free" govts control information so carefully and lie about their statistics so thoroughly that we can use that discrepancy to establish proper weights for our measured lighting scale.
I found myself wondering if it was a lagging indicator. Hopefully the peer review process would have flagged these issues if they were serious. I didn't see the venue mentioned though.
https://www.journals.uchicago.edu/doi/10.1086/720458?utm_sou...
(1980 is an arbitrary date, but before the fall of the USSR and thus the explosive growth of the Eastern Europe, and when shots from orbit likely became easy to obtain.)
And once you're in other big cities, this becomes even more true. It's common to see single floor businesses and buildings right in the busiest parts of town.
I agree with your skepticism of the method and it's good to explicitly list these things. But I think the authors of the research would also probably also agree that the method is far from perfect.
> when people move in to dense modern housing and shift to white collar work the model breaks down....more modern lighting is more efficient
These should apply equally to dictatorships and democracies right? Or at least it shouldn't correlate with the dishonesty of the regime so the model can factor it out.
> people increasingly socialize through phones
You still need light for most forms of economic development. I've been to a few places where it's almost completely dark at night and people communicate on phones. But the economic centers, for example where people congregate for night life, have lights on.
Similarly, gains in energy efficiency, such as widespread LED adoption, are global and not limited to any regime type. The same applies to economic transitions from heavy industry to services and behavioral shifts toward indoor or screen-based activity; these are common across modern economies. However, the study finds that the light/GDP mismatch emerges selectively in authoritarian regimes once they pass the income threshold for certain types of foreign aid.
This pattern suggests that the divergence is not driven by modernization effects alone, but rather by systematic incentives to inflate economic data.
Who should people believe, nighttime light data or their lying eyes?
And the fact that no one just assumes that is weird. In general, let’s imagine you had a politician who took power of a country that was recovering, and then by the time they left power their country was a literal pile of rubble and they shot themselves and their family in the head in order to avoid the consequences of their own actions… you’d assume that any positive story about them is probably bullshit. But for some reason the moment it’s Hitler everyone’s got an excuse.
And if someone accidentally killed 6 million of their own citizens we’d naturally all recognize them as one of the worst politicians in human history, but for some reason when they kill 6 million of their own citizens on purpose it’s not a raucous failure that deserves endless ridicule.
I think that’s an interesting question with probably no simple correct answer.
The same is true all the way back to the Ancient Empires which were also usually ruled by dictators. The era of Marcus Aurelius was an absolute Golden Age in Rome. Yet the era of his son all but ensured the collapse of Rome. Of course the same is becoming increasingly true of democracies where political messaging has become effective enough to regularly make people vote in highly irrational ways.
[1] - https://search.brave.com/search?q=90s+in+russia
Honestly, regression to the mean is a stronger explanation here than "benevolent and effective dictatorship".
I'm flabbergasted that you look at 2025 Russia and consider the word undisputed apt. How ... narrow-minded.
And I think Ukraine is obviously going to be a major turning point in history. Ukraine created an absolutely massive army by combining massive scale forced conscription alongside preventing men of "fighting age" (18-60) from leaving the country, and they're similarly being armed with hundreds of billions of dollars in Western arms - far more than we even supported the USSR with during WW2. And yet Russia, a country that could barely hold itself together in the 90s, and is under severe sanctions, is winning. We're looking at the absolute end of any concept of a unipolar world, and I think that's a great thing for everybody.
[1] - https://tradingeconomics.com/russia/wage-growth
They have a very shit time projecting power across their border, let alone across the ocean.
1. collecting jobs data is a fuzzy thing and guesses are made
2. reporting jobs data is subject to political influence
For example, one “free country” has been arresting people for Facebook posts, and praying silently outside clinics.
US is country #57 (about 25th percentile, from the top) with a score of 84/100 for its general "Freedom of the World" score.
And number #13 (roughly 7% percentile, from the top) with 76/100 for Internet freedom.
[0] https://freedomhouse.org/country/scores
Can’t question western “democracies.”
Nvm it got dismantled by various people who use nightlight metrics for living. NBER also did a sophisticated night lights study of PRC in 2017 - same year as OG Martinez study using naive methodology as Henderson original work from early 2010s - and found PRC's reported GDP was actually _underestimated_ using night light data. But useful idiots will gobble up authoritarian gov inflate narrative.
For obvious reasons this study by multiple staff economists at the FED doesn't get any attention in western MSM, but one by an assistant prof of public policy somehow does. Also note scope, multiple staff economics focusing on PRC, versus assist prof who managed to collate data for every country. Almost as if it's good self marketing considering Martinez keeps updating study with same rudimentary methodology every few years to get clicks.
E: I'm guessing this is now going to make rounds with the Shih/Elkobi's 1/3 of PRC local gov spending all their revenue on debt repayment wank graph being popularized recently that's comparably retarded.
Yet every year since 2017 the Martinez study gets play in mainstream media (MSM) because autocracy incompetent vs democracy plays well. In 2022 Martinez updated the study (I think also 2019, 2024 i.e. milked forever), with same methodology issue said insisted trend continues. Cycle repeats usually synced to some other retarded reporting to PRC economy going to explode (will also be milked forever).
In this case I'm assuming the chart making it's way around reddit and now socials on Shih/Elkobi study that inflate PRC subnational debt situation by conflating principle+interest (tldr bad methodology using non-standard metric to exaggerate debt risk), which like Martinez has been called out back in 2022 after publication. Customary in these discussions of unserious providence is bringing up the unserious Martinez study, which useful idiots continue to take seriously.