New Economics
I am tired of the classic left/right perspective on society, policy, and politics. I realize that markets are an incredible tool to allocate resources efficiently. And I also realize that markets are subject to failure and we need to protect our society from these market failures. I am not a purist on either side of this debate and I find the hard core advocates on the far left and the far right impossible to take. I believe orthodoxy is one of the worst human traits.
So I enjoyed reading this post on “new economics” and I particularly like this table that shows the difference between traditional economic thinking and new economic thinking:
The post goes on to explore how these new economics thinking will eventually impact politics, policy, and society at large.
One particular example reminds me of our work at USV on “Regulation 2.0”:
First, rather than predict we should experiment. Policymaking often starts with an engineering perspective – there is a problem and government should fix it. For example, we need to get student mathematics test scores up, we need to reduce traffic congestion, or we need to prevent financial fraud. Policy wonks design some rational solution, it goes through the political meat grinder, whatever emerges is implemented (often poorly), unintended consequences occur, and then – whether it works or not – it gets locked in for a long time. An alternative approach is to create a portfolio of small-scale experiments trying a variety of solutions, see which ones work, scale-up the ones that are working, and eliminate the ones that are not. Such an evolutionary approach recognises the complexity of social-economic systems, the difficulty of predicting what solutions will work in advance and difficulties in real-world implementation. Failures then happen on a small scale and become opportunities to learn rather than hard to reverse policy disasters. It won’t eliminate the distortions of politics. But the current process forces politicians to choose from competing forecasts about what will and won’t work put forward by competing interest groups – since it is hard to judge which forecast is right it is not surprising they simply choose the more powerful interest group. An evolutionary approach at least gives them an option of choosing what has been shown to actually work.
I also like this bit about the tolerance for failure:
A major challenge for these more adaptive approaches to policy is the political difficulty of failure. Learning from a portfolio of experiments necessitates that some experiments will fail. Evolution is a highly innovative, but inherently wasteful process – many options are often tried before the right one is discovered. Yet politicians are held to an impossibly high standard, where any failure, large or small, can be used to call into question their entire record.
I don’t think the way we do things in startup land should be a model for how everything should work, but there is a lot to be learned from the way the tech sector works. Innovating, trying new things, measuring the impacts of these new things, and evolving leads to forward progress. And when something fails, we accept is as a lesson learned, not something to be embarrassed about (or fired for).
If the worlds of economics and politics are moving in our direction, I am very pleased and optimistic about that.
Comments (Archived):
Governments become self serving monopolies. Policy experiments are an unnecessary risk?p.s. ‘democracy’ and ‘government’ – they have become antonyms.
I like the idea of small scale experiments, but the point about the political difficulty with failure is such a valid point. I don’t know how a politician could survive by proposing we try a little of everything* – but would be refreshing to see that happen.*waffler, indecisive, who’s side are you on?, etc.
We have 50 states….maybe allow them to have some more power to experiment. How to do it without negatively affecting others is tough.
England has 48 counties, but Whitehall in London holds the purse strings of expenditure. There is little policy experimentation.
UK government are increasingly “transferring the risks” to county councils the latest is to raise local council tax to pay for the elderly home care/social services
At the city level, mayors have a lot of power, no?
They do, but issues in cities don’t always replicate suburbs/rural. They have different challenges.
I think states already do their own thing. I don’t know if they experiment by trying many different approaches to solving a problem, but they definitely aren’t bound to just one uniform way across all states.
This was the at the core/initial formation of the 2 parties…one wanted more federal power/control and one wanted more state power/control…but in reality, states are just mini versions of the country as a whole…and it goes down to counties, and then cities…politically, they are all just mini versions of each other. So you could start at the top (more expensive & challenging, but wider reach with risk more spread out) or you can start at the bottom (cheaper, and prob. a little easier to get going but bigger ind. risk and harder to scale & spread).At least in theory, the big guys have the luxury to fail at small things (but often don’t see the long term benefit of doing so in search of the 1% big wins)…the small guys can’t survive failing at anything (but in the startup world this works out because there is an endless supply of ‘small guys’ — so 99.9% can/do die because of a single failure of some sort and yet innovation still happens).I’m not sure how you really work the ‘endless supply’ into politics…but it does feel a bit like the holy grail if someone figures it out!
Yes, I think using either political party as an example is not good. Both parties can be very idiotic. How much failure in public policy are we willing to tolerate?
I’m with you 100%.Everyone is willing to tolerate some level of public policy failure…as long as it’s not in their backyard…and there in lies the *real* problem.
The problems come when you try to quantify anything into a mathematical model. For example, even if innovation is evolutionary, it’s still mysterious because it is impossible to quantify it. I think there is a big divide between Keynesians (salt water) and Classical (fresh water) economists. It manifests itself into a lot of policy.The points about networks is interesting to debate. The price system does coordinate across supply chains. Do non-price interactions matter? Or, are they built into the price? For example, Fama’s efficient market hypothesis says all known public information is built into the stock price that we see. Do non-price interactions also get built into that stock price? How? How can we quantify them?I think that Thaler and the Behaviorists have come across some really provocative stuff. Much of it is very difficult to build into a predictive model. Much of it is extremely scary if we take it to an extreme in implementation.
Isn’t innovation highly quantifiable? You just need to set a goal metric and track it across different policies? For example, how does the high school graduation rate change in districts with and without charter schools? Granted, you can’t model the effects in advance, but that’s a defining quality of experiments.It also seems clear that non-price interactions like network effects matter, and they often don’t get built into the price, but rather into volume. Amazon merchants set roughly similar prices to merchants elsewhere, but because of Amazon’s massive scale, their merchants sell higher volume compared to a small marketplace.
It’s quantifiable looking in the rear view mirror. How do I quantify the ability to implant pig kidney’s into humans replacing organ donation? How do I quantify the utility of a robot? It’s extremely imperfect. As long as the inputs to how you solved the problem are transparent, at least then it can be duplicated. Most of the arguments I see in economics aren’t with outcomes, it’s in the assumptions. For example, a question John Cochrane (The Grumpy Economist) always asks is “Where did government get the dollar to spend?”What I believe is principles matter more than absolutes. If you use basic principles, they can be replicated over and over again. The fewer assumptions behind those principles (and the more transparent) the better. So, if I assume that “people make rational decisions” even though I know full well that some people don’t; I can set up a policy based on that. I don’t know a mom or dad that doesn’t want something better for their children given two choices.This is why a principle that “markets allocate resources better than centralized bureaucracies” is a good one to follow for public policy. Even though we know that sometimes markets do fail. They get out of line. However, centralized bureaucracies fail more. When bureaucracies get out of line, it’s often impossible to arbitrage them so they become efficient.
I see what you mean. Yeah, I think experimentation requires a certain amount of humility about what you can predict ahead of time, which is in short supply in people in positions of power.The problem with principles is that there’s usually a countervailing principle (or many). Markets do allocate resources better than bureaucracies, except for example in cases of externalities or public goods. Depending on which clause you think applies in a specific situation, you end up with completely different positions. In our fairly partisan society, everyone has different priors. If we focused more on experiments and less on principles, I think we’d have a better and more defensible fact base to use for policy making.
isn’t volume a function of price in part
You can build evolutionary computer programs, which inherently requires modeling
If you take the US federal political system as an example, where there are only 2 party choices, it’s almost impossible not to succumb to left vs. right labellings. There is no “center” party to choose from, despite some transient central tendencies that either parties exhibit to win extra votes. That said, the place where centrist and “new economics” influences can do well is at the state, regulatory agencies and local levels. For eg, Mayor Bloomberg was definitely a centrist, economically logical type of person. Regulators should also be in the middle ground of the debate and stay the course on non-partisan policies. That’s why having more than 2 political parties has its advantages, as you avoid the polarization debate that seems to be the starting point otherwise. Personally, I’m optimistic that new economics can win as well, as I find myself most often as a centrist kind of person. More than a modicum of “left” or “right” positions make me cringe, daily.
How can you be a centrist on human rights?
That’s the rub – on any given issue there are basically two sides (for or against).The problem to me seems to be that the “parties” have packaged up all the issues into one set of beliefs – that people now seem to be all-in or all-out on just based on which party they identify with.I’m out on both parties. I’m in on individual decisions…so I guess that puts me somewhere in the center (def. dis-liking both parties and their packaged ideals/options).But this means I often find myself 1/2 agreeing with both parties on a lot of the issues…and prob. 2/3rds disagreeing with both parties on most issues…you figure out how that math ads up (I’m still unable to do so, which probably explains my constant exasperation)
Could not express my feelings better.
I didn’t say that, specifically. Of course I will take extreme positions as needed. But on the whole I tend to gravitate towards balanced views when it comes to economic implications.
of course.this idea of separating your views on economics from your views on societies and government’s responsibility towards our culture is to me abstract by definition.
Another POV might be to pose the question that ‘efficiency’ (of markets) might be a tyranny, in that it does not serve the human condition well.
Very interesting post, thank you Fred. A happy holiday to you and yours.
Applying lean startup methodologies to government initiatives would likely lead to better programs over time. The issue for government, as it is for all other not-for-profit entities, is that the public won’t tolerate anyone experimenting with their money. Innovation requires investment and failure, neither of which are perceived by charitable donors or taxpayers (i.e. most people) to be good uses of their money. I’m not saying it shouldn’t be attempted – just that this is the obstacle to overcome.
I think beyond experimenting with their money, the fear of experiments comes from a deeper place. For example, experimenting with educational methods may be the right thing to do for society as a whole, but I’m sure some of the parents will be up in arms complaining about why THEIR child should be the test subject – understandably so.
Having worked with government for a while now, my sense is that government itself is not as averse to experimentation as people think. In my day to day work at my new employer I’m already working on a project involving matching underprivileged youth with summer jobs, my predecessor helped speed up this process by developing an algorithm to do the matching instead of having people do it. Meanwhile the data scientists in Boston’s Office of New Urban Mechanics are building machine learning classifiers for incoming 311 requests so they can be routed to the correct departments without human intervention. Massachusetts implemented what essentially was a beta version of Obamacare years before it went national. The non-profits giving grants in this world are currently excited about “pay for performance” and other models that involve randomized control trials to test policy interventions.The situation may well be different in other places. However we do not need every place to be a center of experimentation. Boise, ID can easily benefit from the experience of Boston, MA thanks to the fact that we are connected via the Internet. Often times the larger challenges involve getting buy-in to fund or implement things that we already know work. Another is a talent issue. Even if the folks in the middle of Wyoming want to implement a bunch of cool apps and machine learning strategies, they have to convince people who know how to do that to live there.
do you think there are some experiments/types of programs that only make sense at scale when it comes to government intervention?
It’s going to depend on the type of intervention. Anything that basically equates to giving or spending money to do something is probably fine to start small scale and then move up. Modifications to the tax code or other changes that can be circumvented by changing jurisdictions will likely only work well if you implement nationally.
Good post. I have been thinking about the Joi Ito comments ref media echo chambers which is a danger and one you have unwittingly picked up on:https://joi.ito.com/weblog/…
Yes! I wish more people would focus on public policy that actually works instead of whether their side wins or loses. Politics is frustratingly tribal. Beyond the disincentive to fail, there’s also a disincentive to even recognize successes that don’t fit your priors. Instead, parties hide the ball and come up with some other framing for success.Another challenge is that differences in scale can mean that local experiments can act fundamentally different than national-scale policy changes. A state or city adopting a new healthcare insurance system doesn’t change the overall system, but a federal change does.The next four years seem like they will bring a lot more “experimentation.” While that prospect mostly makes me cringe, I hope that there are a few unexpected successes mixed in with the failures. I especially hope the failures are more or less reversible.
Please take more days off. Great post.When Greenspan was hauled back before congress in Nov.08 to account for the crash, he basically buried the notion that markets are efficient and will ultimately act in their (ours?) best interest. The density of rhetoric may keep us from trying new things but eventually reason will return to the economic/political process. Some politicians may come back to the notion of accomplishing something.
Interesting conversation, but as so often with politics and policies; there is only the ethically right or wrong. There is no middle. No room for experiments. This would be as if torture has a scale of acceptance. After 2500 years of ‘experimenting’ with democracy we should have figured out how things need to be setup in order for society to work. https://global.oup.com/acad…
I think your position is the unethical one. In a world where we can engage in experiments to ascertain the impact of new policies, it is negligence to not test them before inflicting them upon the masses.
I wonder, how many more (economic) ‘experiments’ humanity can tolerate. If TRUMP or PRIVATISATION is one these experiments, when and how will you pivot?
Obamacare includes a “test kitchen” to test out various payment options for healthcare services. Although, at least according to this piece, the way it’s discussed by partisans is a bit worrisome. http://www.npr.org/sections…
I really like this. One of the things that I always have railed against is that people are rational.You see this a ton in economics. The efficient market theory for instance. It’s not efficient. I could never get my mind around this.Also it is so right that a small change can have huge differences. It is not linear. For instance the housing bubble. When there are 11 people chasing 10 deals pricing skyrockets somebody will lose out. When there are 9 people chasing 10 deals pricing plummets.Twain Twain had a great Einstein quote that summed it up. “Not everything that counts can be counted, and not everything that can be counted counts.
One only has to look at the market starting at 5pm pacific on Election Day to present to see proof of what you write here!
I just could never get my mind right on this. Chipotle has a food poisoning scare, or somebody thinks Twitter is going to get bought. I am not singling out these stocks just pointing that it is not efficient.
We now have a Pres that engages in market manipulation on a micro level.
One could make the case for that but I don’t want to go political. Let’s move it back to this blog. For instance was it rational for Fred to start this blog? Is it rational to comment? Well there are a ton of nuances and time frames to that.I’d say at this point very rational but during the first year??? Maybe one could view irrational.There are tons of irrational decisions I make every day. The only people that think they don’t are delusional.I’ll give another example. I go to all inclusive (the only thing they really have in the Dominican Republic) before Christmas.I bring $200 in $1 bills and have me and my family hand them out.Is that rational? Well my German counterparts certainly don’t think so. But I like the Christmas spirit and the service and attitude my family gets make it infinitely rational to me.
Went to DR several years ago and stayed in an all inclusive resort around La Romana. I remember driving from Santo Domingo through some of the most impoverished areas I’ve ever seen. Very sad. Your contributions are very thoughtful. My friend does the same daily in NYC. Pocket full of ones.
We always stay at Bavaro, I’ve thought about the Caribbean side. I certainly have seen some poor people when we dune buggy or horseback ride but what we call country poor.I won’t give if you are looking for a handout.But if you are cleaning my room, serving my food and drink, driving, getting my beach spot ready, I feel morally obligated to give.The only thing that gives me a twinge is how so little is appreciated so much. As I walk up I hear “BIG POPPI” (I don’t care what team you like that guy is loved down there) from the men or a song about Sugar from the women. I know enough German and French to know what they think about me, but I don’t care, adding $40/day to my vacation is really not much.Funny story, first time I went to get the ones my bank PNC who’s local branch is so small the tellers don’t require an account number or ID and knows how much I travel really looked at me strange. Ever since I send my wife.
For instance was it rational for Fred to start this blog? Is it rational to comment?It’s rational because both the blogger or the commenter has a benefit from what they are doing.I bring $200 in $1 bills and have me and my family hand them out.Is that rational? Well my German counterparts certainly don’t think so.A guess is that one of the reasons, if not the major reason, you give out the bills is because it gives you a good feeling to do so. You may rationalize it as something else and of course that could be part of the reason but one of the main drives has to be that it makes you feel as if you are a good person to act like that. This is separate from tipping the bellhop which is avoid a negative many times in your brain from no tipping when you are supposed to. Of course I am not saying that you can’t tip because you feel bad for the bellhop or think they are underpaid or because it’s convention to do so. But think of how many examples where you don’t tip.Reasons people tip: Avoid a negative, convention, feel positive from doing so, really happy with service or outcome. A manic tip. Like when the repair guy at the hotel lent me an important cable I needed.Back a long time ago my dad tried to give a tip to a doctor that was treating him in the hospital. That was because he was truly thankful. The doctor? He rejected the tip and if I remember the story was offended. The girls in the management office of the condo I am at one of them was glad to take the tip that I gave last year (the receptionist) the other one I felt was offended. She literally said ‘oh not for me for Linda it’s fine’. Because I think she saw herself as a professional and the act of tipping was for service grade people. In my case I would always be happy to get extra money from someone never going to reject that.
I think you miss a HUGE reason. You will get better service. MUCH better service.Sure if it’s a one time transaction like the bellman it doesn’t make a difference, but if you go there often or are staying somewhere for a while, it makes a really big difference.The word gets out and it makes a major difference in every area of your experience, and isn’t that what you are buying when you go out? The experience.It’s not only the wait time. (I’ll get the I’m glad you came back on time as I just walk up to get to the front of the line) It’s what food and drink you get (they make sure yours is the best), and even the recommendations (mmm, maybe not this, or have you tried that?)
I tip well too. Also look the service workers in the eye and say thank you. Boutique hotel I stay in, in Chicago, was there frequently for a while during a long assignment but now maybe once or twice a year. I am still recognized and receive special attention. Of course the number of people who recognize me is diminishing due to attrition.
I have the blessing of having a very, very easy last name, I also because of my features look like a teddy bear or big baby (my head size is off the charts, I have a pug nose, and I smile a lot), nobody seems afraid to touch me.As you can see from my comments I truly love and relate to regular workers. I deal with so many smart (and arrogant) people every day, dealing with regular people is like getting out of an uncomfortable suit and into my shorts, t-shirt, and sandals. They can feel that. I tip very well, I actually talk to not at or down to them.We have fun with my last name. When I travel people are stunned and how many times I am greeted out of the blue, every week.The MN office had a holiday party and I had to go (hate MN in the winter). Unbeknownst to me they had set me up that if we raised $2k at the party for our charity a school in Cambodia that I would ride a mechanical bull, because that was decided to raise the most money (I don’t even report there). Literally some of the workers at the hotel gave money.They gave one person time off to go film me at the Cowboy bar we went to watch me ride I stayed on the 8 seconds plus another 7 before jumping off. A young guy thought it must be easy and did it for two seconds before giving himself a bloody nose and wrenched back as he went head over heels.I got back to the hotel and the staff was howling and screaming. As usual I caused a commotion. The week before it was in Tucson when I walked into the hotel with a Hershey’s shirt and the front desk clerk started singing Sugar, Sugar at the top of her lungs and saying how sweet I was wearing a Hershey’s shirt (my wife took the kids to the light show when I was traveling)
Love this story, Phil, and the others you shared. You know how to cause a stir. 😉
It’s highly rational though since people are betting on what they think others will do when hearing that information. It’s a loop of thinking.
It might be rational it might not be depending on the outcome (is it rational if you are totally wrong?), but it is not perfectly efficient.
Philip:I have zero expertise in this general topic but, that aside, it’s impossible to not consider your parenthetical question: “is it rational if you are totally wrong?”I’d argue that, generally, one *can* be rational AND wrong.You can make a rational decision based upon methodical reasoning about all observable facts … and get a wrong (non-desirable) outcome.It can be rationale and not generate a right outcome.It’s not solely the outcome which makes something rational.(Conversely, many people have been befuddled by non-rational methods of getting a positive/desired outcome via winning an office pool during March Madness.)
Great point. I have watched two office pool non rational wins: One the person picked NFL teams with animals as mascots, the other was March Madness where the schools picked were the ones the person would have preferred to go to.
trueand this is more common than one would like to admit
It’s rational if you have a reason for your behavior. It’s not rational if someone acts irrationally in making a decision based on emotion and does not have a basis other than that.For example:Rational: I think that the market will go up as a result of “x” event because in the past that is the way that people have behaved. (Example: In the past war makes the market go up/down)Not Rational: I think things will be bad and as a result the market will go down. (Example: The country will go to hell if such and such is elected because I view that candidate as a bad choice).
“It’s a loop of thinking” …that is exploited by fake news and deliberate propaganda ops. Traditional macro doesn’t even work between informed experts, never mind the targets of Bannon’s and Putin’s manipulations.
Gosh, why can’t we stay out of politics?
The assumption the people make rational choices has always been a real sticking point for me too.
Thanks, Phil. I rail against the idea of humans as series of linear, binary rationalities because, well, our species is also about art, culture, social relationships, emotions, empathy, language and more. All of those human phenomena are INTERCONNECTED & INTERSECTING WAVES, not logic lines from A-Z.Recalling another Einstein quote: “Logic will get you from A to B. Imagination will take you everywhere.”Here’s the argument against pretty much every model and system for us that’s been thought up since the Enlightenment, from economics to the tech sector’s models of AI as emulation/simulation of us, our intelligence, our language and our economic behaviors: We’re RELATIVE not rational. We’re PERCEPTUAL not probabilistic. Our intelligence is INTEGRATED not the sum of separated parts. And we’re SUBJECTIVELY BIASED not stochastically biased like the dice or the cards that Pascal & de Fermat used to model “fairness” and “bias” in 1654 with and that (wrongly) got generalized over to model fairness and bias in human beings.What the Enlightenment led to are systems today that are rational, probabilistic, separatist and stochastically biased. So it’s no wonder they can’t understand us! That’s why they’re not “fit for the purpose of modeling us, our language and our values.” They may be fine for modeling the movement of ball bearings (per Newton) but not the mind and motivations of humans.What @fredwilson:disqus defines as “classic left/right” is the Rationalists’ approach to economics we’ve inherited from the days of Adam Smith. In the case of the Rationalists, they called it “invalid/valid,” “illogical/logical” and tech knew it as “thumbs up (like) / thumbs down (dislike)” until it was decided that “thumbs up-only” produced steeper power law, hockey-stick metrics that investors like to pattern recognize.Again, Einstein’s quote about what counts applies because, in power law mechanics, the systems can only count part of the signal (the popularity count that’s the basis of “Trending Topics”). The systems are BLIND to the other part: the “black swan” qualia that could be the difference between one candidate being elected instead of the other or one product being bought instead of another.The blindness thing means the systems aren’t very effective or efficient at risk management because whole spheres of signals are going unmeasured and undetected. That means increased uncertainty. That’s how global financial crisis happens. The focus goes on the rational equations and not the irrationality of how people INTERPRET THE LANGUAGE in the analysis that accompanies the Fed changing interest rates on mortgage repayments.Beinhocker’s “new economics” model is deeply flawed because it’s basically a slight variation on the Scientific Rationalist approach to economics without him even connecting the dots that he’s fallen into the blind spots and traps of “traditional economics”.Take, for example, his point on individuals “Use both inductive and deductive reasoning … subject to errors …” — please see slide 1 because if that’s not Descartes and traditional economics I don’t know what is![This is why Einstein, Da Vinci and Ada Lovelace are smarter than the giants of Rational Science (Descartes, Leibniz, Bayes etc) and today’s economic and technology thinkers any day of the week.]For us to get to more comprehensive and coherent models of economics, it’s vital we know how we arrived at the current dogma (orthodoxy) on rationality, quanta and how we model people, our intelligence, our economic behaviors and our language as mere rational numbers.We have to re-examine the Enlightenment of C17th and the prevailing philosophies at the time.Please note in Slide 1 the inductive and deductive reasoning frameworked by Descartes. A process that, in today’s AI and Deep Learning, we know as “forward-backward propagation”. Note also that in NatLangProcessing the objective function is all about error reduction by probabilistic and statistical means.This is one of the key reasons the MACHINES CAN’T UNDERSTAND US!!!So … Beinhocker’s proposal is looping traditional economics and the same-old-same-old maths frameworks that have created gridlock in economics, boom and bust policies and incoherency in Natural Language Processing. And he isn’t even aware he’s fallen into those same-old-same-old blind spots and traps. https://uploads.disquscdn.c… https://uploads.disquscdn.c… https://uploads.disquscdn.c… https://uploads.disquscdn.c…https://uploads.disquscdn.c…NONE of the leading economists or tech thinkers are aware of how the “Enlightenment” created blind spots for our generation: everything from how we model our human experiences (including economics) to every data+AI system. https://uploads.disquscdn.c… https://uploads.disquscdn.c…Here’s why I base my system on Da Vinci, I Ching, Einstein, Ada Lovelace and not on Descartes’ logic and Bayes’ probability trees.The Rational Scientists of the Enlightenment shifted our paradigms AWAY FROM THE INTEGRATED, WHOLE, COHERENT MODEL OF US. The model where rationality-objectivity coexists in synch with irrationality-subjectivity.https://uploads.disquscdn.c…And that integrated, whole, coherent model is what we need to return to if we’re to have a truly new economics model.So first … someone would have to invent a tool (from first principles) that’s MORE POWERFUL, NATURAL & COMPREHENSIVE than linearly rational binary 0-1 and its probability shades of gray.@wmoug:disqus @JimHirshfield:disqus @MsPseudolus:disqus @le_on_avc:disqus @myscrawl:disqus @donnawhite:disqus @SixgillBlog:disqus @pointsnfigures:disqus @SaltShaker1:disqus @bsoist:disqus @jasonpwright:disqus — Da Vinci’s model for us is way better than Descartes’, clearly.Today is Boxing Day so I was out+about shopping in the sales. On the walk back, I was thinking about leaving tech. Even high-flying women like Regina Duggan have thought of this and little ants like me do too:https://uploads.disquscdn.c…But, see, if I don’t stay in tech, speak up, share and ship a system that shifts the paradigm back to a more coherent model for us (the I Ching-Da Vincian model) … There’ll be another 450+ years of Descartes et al blinding us from the fact that our “irrationality” (our hearts) is as valuable as our “rationality” to our language understanding, our decision-making, our economics, our risk management and our ability to qualify+quantify ourselves and the world around us.The truth is that qualifying+quantifying, rationality+irrationality, objectivity+subjectivity, art+science co-exist in synch in us.That truth is in Da Vinci’s ‘Vitruvian Man’ and in the Yin+Yang referenced in the ‘I Ching’ (the Book of Changes).It’s also in Einstein-Schrodinger’s theory of the cat being both dead AND alive, at the same time. Again, notice that in Da Vinci’s ‘Vitruvian Man’ the logic box and the language sphere co-exists, at the same time.
Every person that has been in tech has thought of just going onto something easier. You have seen my comments, one of my favorite ones is a colleague that says: “you know I think I just want to be a postman in Greece”Keep up the fight.
Thanks for the words of encouragement, Phil.As I shared before, regardless of who won Election 2016, these very deep-rooted problems that persist across all systems means we’re not being well-served.And we need to and can change those wrong paradigms of Descartes that puts us into the left vs right logic trees in the first place. https://uploads.disquscdn.c…@fredwilson:disqus @wmoug:disqus — Every decade or so for the last 200 years that Beinhocker refers to, there’s an existential crisis in economic theory.The persistent problem is that economists write lots of words about Classical, Neoclassical, Keynesian, Friedman, Kahneman, evolutionary etc approaches.YET NOT A SINGLE ECONOMIST HAS BEEN ABLE TO FRAMEWORK THE MATHEMATICAL TOOLS AND EXPERIMENTS FOR HOW WE’D BE ABLE TO MEASURE AND MODEL THE RATIONAL+IRRATIONAL, OBJECTIVE+SUBJECTIVE, LOGICAL+LINGUISTIC phenomena of our intelligence, our decisions and our values.In our more recent time windows of experience …1998: LTCM hedge fund collapse which put at risk the global financial system. It was all about applying rationality, quant and AI to model economics.2008: Global financial crisis. Ditto the trading systems were all about rationality, quant and AI.2016: Bridgewater Capital, world’s largest hedge fund, is applying rationality, quant and AI to model economics. Ditto every other hedge fund platform.2018: ???Regardless of whether it’s politics, economics or technology, if we can’t see and solve for the fundamental flaws in the systems — the flaws that assume rationality and orientates systems design&engineering towards logic objectives with no visibility and acknowledgement of the irrationality (e.g. emotions) that also helps us risk manage information, understand language and relate to the world around us …Then every decade we’ll see and experience the same “wash, rinse and repeat” of value destruction.And the sentiment (emotion) tools AI has … SUCK. There are many tools wrong in tech’s ecosystem itself because of the orthodoxy we’ve inherited and we haven’t even been aware we’ve been applying that orthodoxy of Descartes’ logic and Bayes’ probability ourselves.But now we’ve identified the common blind spots and root problems across data, politics, economics, the sciences and technology-AI.We know those tools of Descartes’ logic and Bayes’ probability are not a good representation or fit for measuring and modeling us. They haven’t been for 450+ years which is even longer than the 200+ years of economics as a discipline.So we move forward into 2017 with that knowledge.
When you get down, watch this video, it always makes me feel better:https://www.youtube.com/wat…
See the “crazy” fellow? That’s us.* https://www.youtube.com/wat…
Is that literally you??? I have a great video of me starting a ruckus at the Rugby 7 championships with another big guy as we started dancing.
Haha, I’m all of 5ft-ish as William, Jim and Shana know because they’ve met me IRL.https://uploads.disquscdn.c….
Ok, that is funny. I have a sister that is not five foot. We would be polar opposites in appearance. I have a 52 inch chest, a head that is so big I literally cannot wear any normal baseball cap, and a 19 1/2 inch neck. My body is not hairy but my beard grows in full in less than a week. They refer to me as SugarBear.
Skinny Minnie me in my 20s. I still fit into those clothes.https://uploads.disquscdn.c…
yeah, I tower over you.
so what is the best way to model emergent systems that are not always rational
Well, it wouldn’t be with the evolutionary genetic algorithms that the computers do and that are considered “hot” investments.Algos that are at the basis of the sequence-to-sequence and Generative Adversarial Networks approach in AI, as used in Google Translate’s sequence-to-sequence and Generative Adversarial Networks. https://uploads.disquscdn.c… https://uploads.disquscdn.c…Notice Pedro Domingos’ comments here:https://uploads.disquscdn.c…The problem is that binary (0,1) is the DNA of the computer’s mind but NOT the DNA of our minds.And that’s at the root of WHY we haven’t been able to model emergent systems like language, economics and culture; coherently.Remember, I rail against Leibniz’s ideas of determinism (0 = there is nothing, 1 = there is something) which Descartes’ logic tree branches as 0 = irrational, 1 = rational (with separation of mind from body from emotions) and then Bayes came along and made that branch into probabilities between 0 = 0% chance of something existing to 1 = 100% of something existing.Blockchain continues with this line of orthodoxy. https://uploads.disquscdn.c…NONE of these Scientific Rationality tools are the right ones to model emergent systems like language, economics and culture.The reason none of those tools are right is that our subjective perceptions are closely intertwined in the DNA of our minds when we engage in language, economics (decision-making), reputation+trust values and culture.0, 1 and % can’t model those subjective perceptions.As man-made tools, they’re not fit for that purpose. At the most, they can do simple forms of “pattern recognition” by probability & statistics.Chris Manning, Professor of AI, Stanford, Jan 2016 Deep Learning Summit: “Higher level language is of a different nature to lower level pattern recognition.”Silicon Valley and the AI community haven;t been able to solve NLU because the only tools they have are lower-level pattern recognition: 0, 1 and %.Ditto economists haven’t been able to model human decisions and behaviors because they have lower-level pattern recognition tools: 0, 1 and %.@fredwilson:disqus @wmoug:disqus @philipsugar:disqus — Therein are the opportunities ahead for economics, data, tech-AI.The computers’ DNA of 0 and 1 bits for evolutionary and emergent systems isn’t coherent, insightful and comprehensive enough to do Natural Language Understanding or to support decision-making models because our DNA markers are more than 0 and 1.The Scientific Rationality tools limit us into boxes and Matrixes of left (0) / right (1) maybe with some probability: left = 20% + right = 80% = votes Republican.That’s why I had to invent another set of tools. Or we’ll all be stuck in the same-old-same-old orthodoxy of left = 0, right = 1 and some percentage.And fake news can’t be filtered with 0,1 and % either.”Learn how to see. Everything is connected.” — Leonardo da Vinci.
Have you ever heard the fama joke?
The article referred to is a good overview of a process that has been going on for decades. The challenge is that whatever shortcomings the classical assumptions have (many) the resultant models are simple and hence malleable. Once these simple assumptions are tossed out, e.g. we accept that human beings are not rational welfare optimizers, then the models become enormously more messy. A long time ago Jay Forrester wrote extensively about the behavior of complex systems. We are still learning a good deal about such behavior and the importance of emergent behavior in such systems.
You forgot a row in your table ;)BlockchainTraditional:Pre blockchain discovery. No enlightened thinking about how blockchains could solve scarcity, uneven distribution, and world peace.New:Base any new potential solution on blockchains to grow the food build the houses and cars and educate and provide health care to the masses!
What fascinates me, both about your post, Fred, and Beinhocker’s premise is that it’s not a methodology that’s required, it’s a philosophy. It’s also not new. As he states:”It should also be emphasised that new economics is not necessarily new. Rather it builds on well-established heterodox traditions in economics such as behavioural economics, institutional economics, evolutionary economics, and studies of economic history, as well as newer streams such as complex systems studies, network theory, and experimental economics.”It’s the use of the word “traditions” that most clarifies what’s required in the “new” new thinking.Much of the methodology attached to the philosophy necessary is also well documented. After all, if you look at the VC model of multiple investments (aka, pilots), each with a plan for scale (aka, PDSA), you’ve got the theory and mechanisms of Dr. W. Edwards Deming (aka, Japanese/Toyota Management/Manufacturing Techniques) – which predates and is the basis for “lean startup”…only broader and always, as Deming contended, just as applicable at the societal level as it is in business.I love this stuff. Even better, it works.
Philosophy is going to be big in the 21st century.As usual, Fred’s a DEW line on it.
Of course, most Americans would describe themselves as “Independent” voters. https://www.washingtonpost….I wouldn’t disagree with what is proposed in “Regulation 2.0”, but I can’t help but feel that our system is already supposed to work that way. Local governments are where these experiments are supposed to be going on (San Francisco is actually a pretty good lab in that regard). There are interesting experiments going on all over the country, but it would be great to see more of that encouraged and facilitated.When I was heavy into democracy activism, I found that people across the political spectrum could agree (among many things) that we need to get big money out of politics. And I can’t help but watch what goes and see it as a spectacle designed to take our eyes off that ball (painting left and right ideas as a zero sum game). Because if the American people ever got together on what we *do* agree on (which in fact, is a lot), the ‘swamp’ would start to drain for real.
The problem is that many of these things that we do agree on are not the same things we think are important or we base our votes on. I’ve knocked thousands of doors across America and the only people that ever cited money in politics as their most important issue were a group of affluent people in New Haven. The candidate who cited that as the primary plank of his campaign lost in both a primary and general election (where he ran as an independent).
That’s a great point and goes to what I’m saying. No one’s going to cite money in politics as their top voting issue, and no one’s going to run on it as their main policy issue, when basic civil rights are at stake (as framed by the two parties).I think there’s a little bridge that gets walked between my ‘most important issue’ and why I cast my vote for the person that I do. And it’s money.Trump supporters will often couch their support of him with caveats about him not being perfect or ideal, but then say they support him because (of their perception that) he’s not accountable to big donors. (That was something people always said about Bloomberg, too.) They feel he’s the only candidate “free” to implement policies in line with their ideologies. If one of the other Republican candidates would have _also_ been seen as free of outside influence (personally wealthy and not taking large donor money), I bet that candidate probably would have won the Republican nomination.One of Obama’s biggest strengths was being perceived as a people’s candidate. The messaging around the volume of his *small* donations was always emphasized. It’s why Hillary had such a serious challenge from Bernie this time — he’s perceived as a ‘peoples’ candidate not indebted to big money.
On Regulation 2.0, do you know if anyone has walked through a specific example or two somewhere? It doesn’t have to be a case study. Would be interesting to read even just a thought exercise that works through a concrete example.
Spot on.
I too am tired of the old right/left tribalism and it is as intense as it ever was. But I find interesting your shot at orthodoxy. Perhaps we define it differently, but I think orthodoxy is natural and necessary.First, without orthodoxy, you don’t have contraianism. Contrarians are simply creating their own orthodoxy based on the world as they see it, and you could even say they “cling” to it until it comes true.Second, it is wildly fashionable to reject orthodoxy these days. It has become an article of absolute faith among the cool kids that orthodoxy is wrongheaded, which in and of itself, IS an orthodoxy.I for one appreciate people who can answer the question “what do you believe to be absolutely true?” and find even more interesting those who can answer “what do you believe to be absolutely true that 90%+ of the world believes to be false?”
I appreciate the people that answer that question like I would which is other than the tenant of treating your fellow human how you would want to be treated, the only absolute truth I believe is that there are no absolute truths. 🙂 Happy holidays.
I’ve had to spell this one out for my boys. For instance, doesn’t mean get your sister the video game for Christmas that YOU would want.
It is a lesson that has to be taught and explained. That just the fact for every child. It’s what breaks my heart the most when I meet people that have never been afforded kindness or had this lesson taught to them. I don’t excuse their behavior or actions or believe they just need a chance. They need the desire to learn and the hard work it takes to overcome that obstacle and I am willing to lend a hand to do that.
Many terrible things come from orhodoxy. The Inquisition. ISIS. Etc Etc.
See my comment, we agree completely when you believe something 100% other than love your neighbor as you love yourself, really bad things can happen. Especially when it is about your fellow woman. Then horrific behavior can be rationalized. Which as I said people are not rational, but they then can rationalize horrific behavior.
dogma.
Important distinction.
Agreed. Many terrible things come from technology, but we don’t reject it wholesale either.
The truly dangerous orthodoxies are those held as a matter of faith for such orthodoxies brook no questioning.
All religion is about control. Sister in law had a baby yesterday. The bris has to be in 8 days. Not 7 days not 12 days 8 days. Except for medical reasons let’s say. It’s about control and keeping people in line with no leeway. The only leeway is that which is granted by some spiritual leader (rabbi etc.). This is meant to give them power to control (since they, like a politician) can grant favors and exceptions etc. so they become more powerful. Not something I read anywhere, something that I have figured out that just makes sense. I am sure some person steeped in religious studies could dispute this but to me the behavioral and common sense aspect always wins out.
Mazel tov. My brother-in-law is a conservative Rabbi and my impression of his Rabbinacal style is to encourage engagement, and not dictate dogma. He does not try to exercise control over how his congregants practice or experience Judaism, and this open-ended approach resulted in higher membership and attendance. How he would wield his power since it hinges on his less than dogmatic stance?
he doesn’t have much power.which is good.
Read or reread The Chosen by Chiam Potok to challenge yourself on this.I just reread most his books. A really great storyteller.
Data-ism is also an orthodoxy, some have argued.https://uploads.disquscdn.c…* https://www.ft.com/content/…
Most good things come from them too: law, scientific research, etc.The current left versus right political battle is tired however.
orthodoxy in reality is complicated. Orthodoxy also gives strong sense of community and values
Second, it is wildly fashionable to reject orthodoxy these days. It has become an article of absolute faith among the cool kids that orthodoxy is wrongheaded, which in and of itself, IS an orthodoxy.This x100. Now Fred is not a “cool kid” 🙂 and I know he thinks things through deeply. That said there is a danger in blanket rejecting orthodoxy — Orthodoxy is how efficiency is folded into evolution. Orthodoxy is how generations pass wisdom to the future. What does change often is the premise and it is absolutely important to understand why something is orthodoxy. That way the orthodoxy can be adjusted as the premise changes. I know we are quibbling on semantics now 🙂 But in a way saying all “orthodoxy is dangerous” reminds me of one of my favorite adage as a product manager “if everything is important than nothing is important”
Maybe it is semantics but according to the definition: “a belief or a way of thinking that is accepted as true or correct”Can cause a huge issue. Since I’m Catholic I’ll take a shot at my religion. People said the Church was always right, that was the orthodox belief. What that meant was they thought they could cover up horrific child molestation and protect the predators that preyed on the innocent.
short answer:That there are truths out there, but we all have poor views of them
Social Finance is a non-profit dedicated to the alternative approach you mention – “experiment , then scale-up what works”. Investors fund “Pay For Success” projects, reimbursed by government after the projects have already succeeded. It’s a very powerful model. http://socialfinance.org/ab…
CPG companies have been engaged w/ product test marketing seemingly forever. The markets selected are considered microcosms of the U.S. based on a slew of different variables, including age, gender, income, household size/ownership, in addition to the competitive makeup of the product category they wish to enter (e.g. brand market share, distribution, pricing). The objective is to minimize risk and the expense of a national rollout, though it’s hardly an exact science as test market success does not always translate nationally. Small cities like Nashville, Cincinnati, Indianapolis, etc. are high on the list of CPG test markets. The notion that economics and policy can be similarly test marketed “cleanly” w/ out agendas and outside influence is a hard nut to swallow. Sadly, the cynic in me says there will always be someone or something w/ an agenda looking to game the system, especially when policy and economic interests–local, state and/or regional–are at stake.
Moderation in all things – including traditional economics
When I woke up hungover at the lake, I will include moderation in this mantra (my Mom’s favourite).
Traditional economics is based on rational man. Human nature being what it is, the market cannot be permitted to allocate goods and services without some regulation.
I believe orthodoxy is one of the worst human traits.Indeed!
Assume a can opener 🙂 https://en.m.wikipedia.org/…
Evonomics seems to constantly claim that “classical” economics says things that I’ve never heard any economist ever say. Economists don’t claim that people are “perfectly rational” or have “perfect information” or that markets achieve “perfect equilibrium.” The simple Marshall Diagrams in an intro economics class might be drawn with that claim, but they are always introduced with the caveat that the real world is more complex. But: that doesn’t mean they’re wrong. In the real world, demand curves *do* slope down. I’ve never heard anybody who has demonstrated that they understand what that means dispute it. (And that one axiom leads to dozens of important results that policy makers routinely try to ignore, typically by trying to legislate prices.)
“we need to get student mathematics test scores up” — Sure, experiments in public schools have been pushed aggressively by tech, hedge fund and now vc moguls who are careful to exempt their own kids sequestered away in private schools. How about we switch things up and start with the experiments in private schools where there’s far less bureaucracy and then replicate the success to public schools?
i find it extremely funny and intriguing that the “new economics” resonates to me as very similar to the unwritten rules/assumptions that underly the business culture and business thought in china!rather than old vs new, i’d say west vs east. or old vs. even older. Nothing is new under the sun.
Interesting that you choose “orthodoxy” as the enemy – I always considered “literal-ism” (granted a close cousin) as the real threat to progress. If there is one thing we have learned over time – its that if it can be committed to writing, it is subject to change and, in most cases, dramatic improvement. Science and innovation embrace this but for some reason, many in government and religion often deny its possibility. Orthodoxy in religions and perhaps constructionists in government? Is it possible that we committed to writing the best form of government ever to be created over 200 years ago? Or shall we dare consider progress in that arena as well.Great post. Got me off the sidelines….Happy healthy New Year to all.Gregg Freishtat
Yes and no. I’m not against experimenting, but I’m always amazed at how a lot of stuff, especially rules and regulations isn’t well thought through to start with.My first job, in sales at Dell, my commissions rose disproportionately the more I sold each quarter (ie, fake figures, but let’s say 5% up to 50% of goal, 10% from 50 to 100% of goal; 15% above 100%). Guess what ? I delayed sales one quarter, and tacked them on to a big push every other quarter. I wasn’t even sly about, I just told my boss (and made sure it was synced with the end of the fiscal year ;-p).Other example, in France “training” gets a tax break, “installation and configuration” doesn’t. All accounting software I sold came with weeks of training, and no installation…To me, what’s really important is to monitor, evaluate, and iterate. We know software always has bugs. Plans and rules too.
>> Economics … can become trapped in sub-optimal states.This is self-evident – none-the-less important to note that policy makers bury their head in the sand (does anyone remember the 99% protests – did anything significant change ?Note :: I am no bleeding heart liberal – but the idea that rewards are allocated justly or in accordance with effort is a) entirely subjective – what is a just allocation ?b) patently nonsenseFinally – does a patent system even belong in a capitalist system ? – Is it not an outmoded notion that was intended to lend short term monopoly rights when 20 years in business was still short term because of communications delays etc?
LOL this “new economics” book only about 30 years late
By far and away that new economics article you linked to is among the best articles I’ve read all year.Their examples have intrinsic problems, such as after the fact predictions versus of the moment, and how well data can be gathered for of the moment predictions, but it is definitely among the best ideas of the year, and worth me spending more time pondering