Author: hunterhawley

In a capitalist society, there are two ways to build wealth: wage labor and capital returns. Despite our ideals, the overwhelming majority of Americans are far more familiar with the former than the latter. While most have some exposure to capital returns, such as through homeownership or retirement accounts, they are often modest. For far too many Americans, these do not constitute a large enough proportion of their income to consider them earnest participants in the ideal of American capitalism.

However, even a staunch anti-capitalist would find it hard to argue that the system isn’t working for those with a much higher-than-average proportion of their income derived from capital returns vs. wage labor. So how do we create this balance for more Americans? How do we ensure that every machinist, nurse, landscaper, and engineer is not solely reliant on their direct labor to support themselves, but also on the capital returns generated by that labor?

Two recent trends offer the greatest opportunity in our country’s history to do just that.

First, the well-documented and impending “Silver Tsunami” of Baby Boomer-owned small businesses looking for buyers suggests that the next decade will present one of the greatest wealth transfer opportunities in US history. Many aging business owners find themselves in the strange position of operating a profitable, attractive business without children or other successors willing or able to take them over. By itself, the Silver Tsunami does not present an opportunity to help America live up to its broad-based capitalist ideals. Instead, it threatens to erode our position even further as wealthy LP-backed private equity firms buy up these small businesses, which employ roughly half of private sector workers.

As for the second trend? A 5-year effort since the passage of the Main Street Employee Ownership Act of 2018 to make good on the bill’s promise to expand access to capital for Employee Stock Ownership Plan acquisitions. In late 2023, this effort produced a truly riveting document from the Small Business Administration (SBA): Standard Operating Procedures 50 10, effective November 15, 2023.

It’s hard to imagine why one wouldn’t be interested in reading this 420-page document, but here are the highlights, just in case:

Employee Stock Ownership Plans (ESOPs) are a popular way of structuring a company so that a broad-base of its employees have an ownership stake in its stock. In some cases, the percentage of such a company’s stock owned by its ESOP — and thus by the employees — is the minority. In other cases, however, 100% of a company can be owned by and fairly allocated among its employees via an ESOP.

As you might imagine, this altered incentive structure works well for employees and for overall company performance. A 2017 study conducted by the National Center for Employee Ownership found being in an ESOP was associated with 92% higher median household net wealth, 33% higher median income from wages, and 53% longer median job tenure. Plus, several studies have shown that when compared to their peers, ESOP companies have a higher return on assets, higher profit margins, and are only 75% as likely to go out of business.

Despite these advantages, only 18% of US private sector workers have any ownership stake in their company, which is where the SBA comes into play.

SBA 7(a) loans are federally-backed and can be used to acquire an existing business. Often, would-be business buyers are unable to secure a loan due to a lack of sufficient collateral or a number of other reasons. In these situations, the SBA backing empowers banks to take more of a chance, knowing that if the borrower defaults, they can recover part of the loan from the US Government. ESOPs are often in this class of would-be buyers that lack sufficient collateral, but until 2023, the SBA 7(a) program was not practically available to them. This meant that if a business owner wanted to sell to their employees via an ESOP, they had to wait years, sometimes up to a decade or more, to be paid for their business. Many business owners simply don’t have the time or financial security to wait, and thus couldn’t consider an ESOP.

However, SOP 50 10 lifted the two major financial roadblocks which effectively barred 100% ESOPs from receiving SBA 7(a) capital. First, the SBA made an exception for ESOPs to the requirement that business acquisitions be made with a 10% equity injection (down payment). Second, they removed the requirement that an owner of the business make a personal guarantee, often putting their primary home on the line, to secure a 7(a) loan for a majority-owned ESOP. Neither of these rules were practical for a 100% ESOP, as ESOPs don’t require cash or personal guarantees from participating employees, and they are the only owners. Now, with SOP 50 10 in place, an incredibly powerful method of financing the formation of ESOPs has become available.

---

Staring down the Silver Tsunami and its wealth-transferring potential, armed with such a powerful new financial tool, there’s no reason why America cannot harness these two converging trends to shape itself into something closer to what it has always sought to be: a land of opportunity for all.

The only question that now remains is whether it will.

It occurs to me that the valuations of many “AI companies” are too high relative to the valuations placed on certain proprietary datasets, many of which are owned, controlled, and maintained by companies whose valuations are driven by more traditional means, but whose businesses also stand to benefit immensely from further AI development due to the datasets that they maintain.

In the realm of AI, tech companies and startups are making substantial investments in research and development, and are receiving a substantial lift in valuation from the potential of the technology. However, they are only able to create solutions for scenarios for which they have sufficient training data. This has constrained AI development to datasets that are publically available or that can be easily manufactured through newly-formed operations, sometimes via labeling services like Amazon Mechanical Turks.

This explains why we are seeing generative AI models do amazing things in language, code, or image generation: There is an abundance of free data online upon which to train a model for these purposes. However, for AI to become useful in many industries, and for AI to deliver a true competitive advantage to any firm, it will require models trained on high-value, proprietary datasets.

On the other side of the landscape there are established firms, often non-tech or non-AI-focused, that possess incredibly valuable datasets that cannot be found online or easily replicated without building the specific businesses they operate. Despite holding the key to creating highly valuable AI models, most of these firms are not directing their investments toward AI R&D, and thus are not experiencing the associated valuation lift.

Specialty insurance, agriculture, and healthcare all sit in this sweet-spot, along with several other industries. Each are especially data-rich, and firms in these fields heavily rely on their data to succeed. Unlike the process for training existing language models like ChatGPT, which take publically-accessible writings from the internet as their training data, you won’t find claim data from insurance providers, yield data from agricultural companies, or patient data from healthcare firms floating around online. Each firm receives a competitive advantage – or disadvantage – based on the quality and quantity of its proprietary data, one that is greatly exaggerated by future AI implementation.

Bowery, a NYC-based vertical farming company that has raised over $700 million, realized the value of agricultural data amid the advance of AI from its start. Alongside vertical farms, they built BoweryOS, an AI-powered system that adjusts the amount of light, water, and nutrients an individual plant receives based on several inputs, including computer-vision analyzed photos of the plants. By playing both sides of the equation, generating massive amounts of proprietary data and developing the AI to maximize its value, Bowery enjoyed valuation multiples that would make a traditional farmer gasp. But as interest rates have risen, and venture capital has dried up, Bowery is demonstrating perfectly the difficulty with developing both the AI and the underlying business operation required to train it. Fidelity has written down their valuation by over 85% from the height in 2021, which is a far better outcome than several of their peers, such as AeroFarms, AppHarvest, and Kalera, who have all filed for bankruptcy this year.

Now, I am personally a big fan of what Bowery and similar firms are building, and despite the setbacks, I still see a bright future for their technology. The lesson in all of this, for me at least, is that there are a class of businesses out there who, due to the operational data they already produce, are wildly undervalued, especially when held next to companies developing only the technology half of the AI equation, or who are trying to tackle all of it at once. It seems far more likely for an established company with great data to leverage AI than it is for a company with great technology to manufacture comparable data, especially through the various economic cycles we’re likely to experience before AI is fully adopted.

In a few years time, a handful or fewer of firms will win big in the race to develop the best AI technology, and they will have undoubtedly spent billions of investor dollars in the process. Meanwhile, there are countless firms who have yet to realize that they are sitting on a goldmine of data which will become exponentially more valuable once AI development is subsidized by venture capital and big tech. In the meantime, the companies developing the tech enjoy inflated valuations, and the companies sitting on the data are being presented with a huge opportunity.

The gig economy is nothing new: In 1960’s Kentucky, my grandfather learned to weld by working “piecemeal” – an arrangement where you’re paid by the number of parts you finish. Even then, this was an old-school, pre-industrialization practice that he was only able to land in the ultra-rural part of the country where he (and I) grew up, outside of major industrial hubs. Today, apps like Uber and DoorDash have brought the piecemeal concept to the digital age as we see more workers and businesses shift away from traditional 9 to 5 work.

Plenty of companies are searching for ways to cater to this newly resurrected form of work, but it’s a hard thing to get right. Worker acquisition costs can spiral out of control, especially if worker retention isn’t high enough to recoup the up-front investment, and the whole thing blows up if you can’t get high-quality, well-trained work from a workforce of non-employees with lots of other things (and gigs) on their minds.

While building Blueprint Stats, we routinely had to scale up and down our “statistician network” – the people who we paid to watch basketball game film and use our software to tag statistical events. Because of the seasonality of that business, I had a unique opportunity to take several cracks at building a worker acquisition, evaluation, and retention machine. In the early attempts, we didn’t get a whole lot right, and often had to sacrifice unit cost and turnaround time in order to keep up the work quality. As we grew though, I was able to hone in on several factors that allowed us to create a successful, international “piecemeal” labor force of over 500, and here are my top 3:

1. Meet workers where they are

The header may sound obvious, but if you want a shot at keeping worker acquisition costs under control, you have to treat the endeavor in the same way that you treat CAC (customer acquisition costs). At Blueprint, we found workers in the same way that many consumer apps do: through online ad placements and crucially, referrals.

Before setting out on a new ad campaign, we would create a persona of the ideal statistician: where they lived, what they interacted with online, and what they were motivated by. Then we worked backwards. We created video ads that messaged to the motivations of our target audience (“watching basketball is cool, and get paid for it is cooler”), placed them before YouTube videos that we knew they were watching, and targeted the countries where the cost of living and interest in basketball were at ideal levels.

Once we had a base of statisticians from ads, we relied heavily on referrals for lower-acquisition-cost, higher-quality workers. To execute, we followed our 50/50 forumla: offer the top 50% of our existing workforce a bonus equal to 50% of our ad-driven WAC (worker acquisition cost) for each statistician that joined, completed their evaluation, and worked 5 assignments. This not only added an incentive for the workers to send a link to their friends, but to help them pass the evaluation and complete 5 assignments – a key metric toward long-term retention.

2. Evaluate fast, and only where necessary

Today’s workforce does not want to wait the 4-8 weeks of a standard hiring process to start working. Ideally, you would have sent them a task yesterday. Additionally, they won’t go through multiple rounds of interviews, several hours of tests, and a number of evaluation tasks before getting paid. I can hear folks screaming at their screens now that “workers expect too much these days” – yes. Get over it, or get left behind. Remember how I said you have to treat gig workers like customers when it comes to acquisition? That carries on throughout the whole process. You have to compete for them.

What this means on a practical basis is that you have to ruthlessly cut out any fat from your evaluation process. If it isn’t strictly necessary that a worker have a competency in a certain area, do not evaluate on it. Additionally, it helps to set up tiers to get workers through evaluations and earning money quicker, and to gamify the experience so that it feels like workers are “moving up the ranks” without the standard promotions that come with employment.

At Blueprint, we had a super-easy evaluation that most statisticians could pass on their first try without reviewing any training materials. While we preferred a statistician have skills outside of the accuracy of their work, that is ALL we evaluated on at this stage. Once they passed, they were able work on our lowest-paying tasks (from signing up to being assigned your first task could take as little as 45 minutes). Eventually, they were offered the opportunity to take a much harder evaluation with more exciting game assignments and better pay on the other side.

3. Pay often, pay fair

I’ll say the quiet part out loud: businesses find the gig economy attractive in part because they don’t have to offer benefits, pay employment taxes, or in some cases, even minimum wage. Plus, it filters out less-productive workers as they aren’t able to complete tasks in the amount of time needed to earn a livable hourly-equivalent (this is a double-edged sword, of course, because it also incentivizes rushing to complete tasks at the expense of quality).

Here’s my take, for what it’s worth: if you have any interest in sustained success, you once again have to treat gig workers like customers, not as disposable labor. Uncompetitive pay is a quick way to generate churn, and churn is your enemy. Lag time on pay is another great churn-generator. Gig workers aren’t employees, and they don’t expect to be paid weeks after a task has been completed. Same-day is quickly becoming the norm, and it’s what we went with at Blueprint.

There are so many benefits to the gig economy beyond the fleeting ability to exploit workers and get cheap labor, and further, between government intervention and the ever-increasing competition to hire gig workers (often from companies that you didn’t even think were your competitors!), if your model relies on that kind of behavior, it’s not a sustainable one.