The government should pay entrepreneurs salaries to save the economy (Trickle Up Stimulus Optimization)
A 2009 Crisis-Era Proposal That Predicted Andrew Yang’s UBI
Note: this is a blog post I originally wrote in 2009 during the financial crisis. In 2019 I became a supporter of Andrew Yang, and remembered that my proposal ten years earlier, while not as “complete,” was very similar. So I updated it a little and surfaced it back. Despite having a degree in Economics, I am NOT an expert on the economy but an expert on behavioral design and gamification.
A few weeks ago, I was exercising while listening to the Wall Street Journal This Morning about what the government is doing to save the economy. I have also been paying attention to how governments give grants to startups that can prove they are highly innovative.
Having written a blog post on this topic earlier, I formulated what I think is a workable plan for the government to save the economy.
For validation, I took this plan to two friends: one an ex-VC and Boston Consulting Group consultant, the other a Stanford University researcher. They have not been able to poke holes in this theory yet, so I am sharing it on the blog and hoping to either find the flaws in my thinking or get it discovered by policy makers who could execute it.
The Foundation: Three Coefficients That Determine Stimulus Effectiveness
When the government injects money into the economy, it passes through many “nodes” (a person or organization), and each node carries three coefficients along with it: spend/save, innovation, and upside. This is a tiny bit of econometric reasoning I picked up from my UCLA economics degree, but it breaks down into fairly simple terms.
The Spend/Save Coefficient
The Spend/Save Ratio measures how much of the received money a node spends vs. saves. If a node saves all of the money once it arrives, the economy will not improve from that injection (unless it later gets invested into a business with a high spend/save ratio).
In times of uncertainty, the wealthy save. The only people who reliably spend are low-income earners who can barely afford their own living. These people are forced to spend essentially all of the money they have each month, boosting the economy forward, especially if they pay other low-income earners in turn.
The Spend/Save Coefficient determines how many “nodes” the money will pass through before it stalls. That node count is what makes the other two coefficients relevant in the first place.
The Innovation Coefficient
The Innovation Coefficient measures how much innovation each dollar produces as it passes through an entity. Like most corporations argue, innovation does not just mean creative ideas — it includes implementation and market adoption.
Some corporations spend hundreds of millions to launch an innovative product. A startup creates a similar (sometimes better) competing solution after spending one million. The startup’s Innovation Coefficient is far higher because each dollar produced more innovation.
The Upside Coefficient
The Upside Coefficient measures how much more money each dollar can generate if the bet pays off. Venture capitalists know this concept well. A VC will fund ten startups that each have a potential upside of $100M. Most fail, but it only takes one success to cover the rest and then some.
If the upside of the same ten businesses were only $10M each, the math collapses. Investing $5M into ten businesses means $50M deployed; only 1-2 companies return $10M each while the rest die. The portfolio loses money.
For government stimulus, the goal must be a high upside value — because the actual objective is to create thousands of jobs for each batch of money deployed into the economy.
Optimize Government Spending by Maximizing All Three Coefficients
Each dollar the government spends needs to pass through as many nodes as possible, with high coefficient values at each one. The formula simplifies to:
[$ × spend ratio × innovation × upside] for however many nodes there are until the money settles.
Most government stimulus plans optimize one coefficient and ignore the other two. Cash to large corporations: high upside, low spend/save (they hoard cash in uncertainty), moderate innovation. Cash to consumers via tax rebates: high spend/save, low innovation, low upside. Bank bailouts: low everything because the money mostly stays put. The challenge is finding a single node category that scores high on all three.
Why Direct Entrepreneur Salaries Maximize All Three
Through this framework, my conclusion is that the best way to pump up an economy is to give grant-like salaries to sustain the lives of entrepreneurs. (Perhaps biased, but I sincerely think this is the best solution.)
The government can give entrepreneurs a deliberately low amount of survival money — say, $20,000 a year — as a salary to do entrepreneurial work. They still have to raise their normal funding from investors and grow their business; this plan only covers their personal survival floor so they can pursue the innovation work without fear of dying on the streets.
Many entrepreneurs are among the smartest and most creative people in the workforce, and they demand very little to survive. They also spend nearly all of that $20,000 because they have no capacity to save it.
I tried this myself. $1,200 a month is enough to cover subletting rent, cell phone bills, food money, $250 worth of monthly gas, and everything else I need. $20,000 a year is already a premium from that standpoint.
That stipend produces 80 to 100 hours a week of productivity. I worked 100 hours a week for years without any pay. If an entrepreneur is running a startup to have a comfortable materialistic life, they are in the wrong profession.
Entrepreneurs would also likely spend on other smaller, cheaper companies and employees — who also need to spend most of their money. Their dollars pass through many nodes, exactly what the spend/save coefficient rewards.
Their innovation value is obviously also very high. Each dollar spent in a startup is more efficient in terms of innovation than the equivalent dollar in a large company, because startups have to do everything the leanest way without bureaucratic overhead.
A great talent working 80 to 100 hours a week on an innovative concept, for $20,000 a year. Am I missing something here, or is this the best deal an economy could ever have?
Finally, and most importantly: if a startup works, it can make millions in revenue and create many more jobs. Startups don’t raise money to make cash flow that sustains their own lives. They raise money on the promise that if it works, the capital put in returns a hundredfold — creating new jobs, new markets, wealthy investors, and new happy entrepreneurs.
The Math: 100 Entrepreneurs for $2 Million
With $2 million, the government could support 100 entrepreneurs for a year. (Not even talking about the billions spent elsewhere saving dying companies.) If you have good VC-grade talent screening which startups have potential, the numbers game says statistically that 10 of those 100 startups become successful and generate millions in revenue, with thousands of jobs created.
Now imagine if the government spent $200 million on this instead. That’s 10,000 entrepreneurs funded. Even at a 10% hit rate, that’s 1,000 successful startups, each potentially creating hundreds to thousands of jobs.
The beauty is that the 90 entrepreneurs who fail haven’t wasted the money either. They produced some innovation, pushed forward valuable skill sets and experience in their fields, and spent all of their stipend into circulation through high-spend/save nodes.
To be clear: it is still up to the entrepreneur to figure out how to raise additional capital, find talent, and grow the business. The stipend covers personal survival only. Everything else they earn on the open market.
This program would also raise the entrepreneurial culture of the area dramatically. Entrepreneurs from other regions would move just to apply. Young talents would pursue this path immediately after college instead of taking corporate jobs to pay rent. Overall, a win-win-win for spending, innovation, and job creation.
Response to the “This Is Just Welfare” Critique
I have received pushback that this proposal is essentially welfare for entrepreneurs, and that welfare is inefficient because the government should be hands-off. I disagree, and the disagreement is structural, not philosophical.
In welfare systems as commonly designed, you pay people who are struggling to survive — and that’s where the spending stops. The money usually settles after one to two nodes (the recipient buying necessities). In my plan, you are giving money to people who could, statistically, create 1,000 more jobs each if their startup succeeds. The money keeps moving through the node chain because of the high upside coefficient.
One is more like a sympathy vote. The other is more like investing into the future of the economy.
I would argue that even if only 1% of the people the government supports on traditional welfare could create 1,000 more jobs through this kind of structured entrepreneur stipend, the economy would be much, much better than it is today.
The 2009 crisis is over, but the underlying argument has only gotten sharper over the intervening years. Andrew Yang’s UBI proposal in 2019 was an even more generous variant of the same idea (universalize the stipend, drop the entrepreneurship gate). The behavioral design question — which design of cash transfer maximizes spend/save × innovation × upside across the most nodes — is still open and still the right question to ask.
Frequently Asked Questions
What is the “Trickle Up Stimulus” idea?
Trickle Up Stimulus is the inverted version of “trickle down” economics. Instead of giving large tax cuts to corporations or the wealthy and hoping the benefit cascades downward to workers, the government directly funds the high-leverage nodes — in this case, entrepreneurs at survival-level stipends — and the benefit cascades upward as those entrepreneurs spend the money into local nodes, build companies, and create jobs that flow back into the broader economy. The model is a node-by-node multiplier, not a top-down dividend.
Why $20,000 a year and not more?
Two reasons. First, $20,000 is a survival floor that genuinely covers the lean-startup lifestyle most early-stage entrepreneurs already live. Second, keeping the stipend at survival level preserves the selection mechanism: entrepreneurs who would do this work regardless are the only ones who self-select in. A higher stipend would attract people who are not actually committed to entrepreneurship and want a comfortable job, which collapses the upside coefficient that makes the plan work.
How is this different from existing small-business grants?
Existing grants typically pay for business expenses (equipment, marketing, contractors). They do not cover the entrepreneur’s personal living costs, which is the actual bottleneck for many would-be founders. Most aspiring entrepreneurs in the lower-middle class cannot leave their day job to start a company because they need rent and food money. A direct salary at the survival floor removes that single blocking constraint without touching the much harder problems of finding product-market fit or raising venture capital.
What is the relationship to Andrew Yang’s UBI proposal?
Yang’s UBI is the universal version: $1,000 a month to every adult American, no eligibility test. The proposal in this post is the gated version: roughly the same survival floor, but only for verified entrepreneurs. Both reach the same destination (a baseline floor that frees people to pursue value-creating work), but UBI is structurally simpler at the cost of much higher total spend. The entrepreneur-gated version costs orders of magnitude less because it only funds the highest-upside-coefficient nodes.
What is the connection to gamification and the Octalysis Framework?
The proposal is behavioral design applied to macroeconomic policy. Core Drive 1 (Epic Meaning and Calling) is what makes someone work 100 hours a week for $20,000 a year on a startup — they are not in it for the money, they are in it for the chance to build something that matters. Core Drive 3 (Empowerment of Creativity) is what makes them more innovative per dollar than a large corporation. The plan only works because entrepreneurs are an extreme tail of intrinsic motivation, and gamification design is the study of exactly that tail.
Has any government actually tried this?
Variants exist in different countries. Singapore’s Startup SG Founder grant, Chile’s Startup Chile program, Estonia’s e-residency-driven startup pipeline, and France’s French Tech Visa each include some form of state-funded entrepreneur support. None of them is structured exactly as proposed here (most have additional gates, equity stakes, or expansion requirements), but the directional evidence suggests the underlying economics works. The 2009 version of this proposal predated several of these programs.