I recently announced a quest to find the most “profitable” places to be self-employed. I am posting my deliberations online in case others might be interested in my thought process and / or findings.

As I started researching various locations, I found that my initial model of “Median Income – Median Rent” was a little too simple to capture the realities of living in one place vs. another. The hard part is estimating how much I’d earn. For instance, some of the wealthiest neighborhoods are sparsely populated, so it’s hard to see how I’d pull in good business there. This has forced me to take a closer look at what I need in a locale. That, in turn, depends very much on the services and clientele that I serve. Before I can get much further, I need to take a closer look at myself and my needs. This analysis will have very different details for each businessperson, but the same principles will apply: Define your market households and find out where they are.

My Services

I run a unique shop as a tutorney: part tutor, part attorney. A look at last year’s bookkeeping reports showed me how last year’s revenue was distributed across three categories of service. The breakdown was pretty close to:

• Legal and patent services = 50%
• One-on-one tutoring = 25%
• Group classes = 25%

Before I looked it up, I didn’t even know this basic fact. I’m learning something already! Thus far, I need all three streams of income to make a living.

My Market Households

To complicate matters, these services cater to different demographics.

My core clientele for legal and patent services is small businesses and mid-to-high income individuals, so let’s say the market household is $100,000+ income. Clients call into my law office from a fairly broad region of the Los Angeles metropolitan area, mostly within, say, a 25-mile zone. It wouldn’t make much of a difference where my law office is located within a city. It’s the city that will make the difference. I need to live in a city with a high number or density of $100,000+ households. (I would rather not adjust this for the cost of living across various cities, because it’s a ballpark figure anyway).

The $100,000+ income bracket is also a good measure of my market household for one-on-one tutoring. Most of them are high-school families. This market is much more localized; most of these clients don’t bother driving more than about four miles to my office. To maximize on this market, it’s best to find a neighborhood within a city that has a high market household density. It’s not very important to have super-rich clients. $500,000 households don’t need five times as much tutoring as $100,000 households. It’s the density of “rich enough” households that matters.

Third, I teach group “test prep” classes to college graduates applying to graduate or professional school. In this market, my clients tend to be low-to-mid income. My classes cost about half as much as the big agencies’ (like Kaplan), so I draw the bargain-hunting crowd. That works for me, because I group them together and earn a good hourly rate. Group-class students are less localized than high-school students. They sometimes drive up to 25 miles to get here, which is practical on the weekends. Living near a college would help if I could really tap into it, but that’s hard to do. Most of my students are already done with college. I’ll say my ideal environment for group classes is a metropolitan area with a high density of $50 – 100K households 1They almost always live with other earners, so this really is “low-to-mid” for 2 or 3 incomes combined. , with nearby colleges being a nice bonus.

It’s probably not possible to find a neighborhood that is optimal for all three of these markets, but I should be able to identify strong combinations. For example, areas with uniform distributions of low-, middle-, and high-income households offer some of all these markets, while also offering low-rent housing. If I had to sacrifice one of these markets for the others to grow, I would give up the one with the lowest hourly rate — high-school tutoring.

Competition

Growing your business is not just about finding a bigger market. You also want a big market share. That depends on the number and size of competing businesses. It’s a major factor that I think many entrepreneurs overlook. A friend of mine is one of the most successful immigration lawyers in Phoenix, AZ. She expanded to LA and counted on an easy-breezy doubling of her fortune. What she hadn’t counted on was the dozens of competing immigration lawyers near her LA office. Dozens — not in her neighborhood but in her freaking building. It was a market she was not prepared for. That office was closed within a few years.

My friend’s example shows that competition might not be proportional to city size. In line with that, the large national test-prep chains like Kaplan can’t be in every city. They focus on the very largest college communities in the very largest cities. I currently live between UCLA and USC, but I hardly ever get students from either school. Blueprint and Manhattan Prep occupy their campuses like invading armies. Maybe a school like Cal State Northridge or the University of Minnesota would be slightly less under their domination.

Toward a metric

If I can capture the most important factors, I can construct a sort of “earnings metric” for each city or neighborhood. From the considerations above, those factors would have to include:

• A = Number of $100,000+ households within a 25-mile radius
• B = Number of $100,000+ households within a 4-mile radius
• C = Number of $50,000 – 100,000 households within a 25-mile radius
• D = Number of competing patent lawyers within a 25-mile radius
• E = Number of competing high-school tutors within a 4-mile radius
• F = Number of competing test-prep centers within a 25-mile radius

Then my earnings metric would be something like:

M = 0.5(A/D) + 0.25(B/E) + 0.25(C/F)

When measured against the baseline neighborhood where I live now, M would be a good first guess for an earnings multiplier. That is, if M = 1 for Rancho Park and M = 2 for Austin, then I could reasonably expect to earn twice as much in Austin as in Rancho Park. To be more precise, each ratio will actually be a ratio of ratios: the new neighborhood ratio over my current ratio. But now we’re getting into formulas that are too hard to type in a simple blog post. 😛

The next phase of the analysis will be determining the baseline ratios for my current location: Who lives here, and who’s serving them in competition against me?

Step 1: Formulating the Problem

I hate moving. I refuse to do it more than once in five years. However, I will relocate for good reason. Now I’ve been in one spot for six years. I am not tied down to an employer and, sadly but importantly, my cat only has a few months to live. In a nutshell, I am completely liberated to move anywhere I want. So … wow, where do I go from here?

I came to Los Angeles two decades ago for law school, and I’ve been here ever since. Most people’s response is, “OMG; LA?! it’s so expensive there!” Now, that’s only half the story, but it’s the half that everyone thinks of (probably because most people are not self-employed but think in terms of fixed salaries). LA is also known as an exceptionally large market with strong earning opportunities. I’d be likely to earn less anywhere else. But so far this is all conjecture. Some locations must be more profitable than others, and I might as well identify the best. It’s not a decision I’d base on word-of-mouth. So now I am on a quest, a data-mining-and-analysis-quest, for the most profitable location to live. I will only undertake a long-distance move if it would clearly offer a more profitable lifestyle than I could find locally.

Note that my operative criterion is “profitable”. The business mandate is to maximize profit: revenues minus expenses. For a person, this translates to:

PERSONAL PROFIT = TAKE HOME PAY – RENT

Modeling notes: This assumes that non-rent expenses would be pretty much constant, i.e. it would cost roughly the same to eat, make car payments, or visit the doctor no matter where I lived, and that rent would dwarf any other single expense anyway. Local or state taxes could have an appreciable impact on take-home pay.

Take-Home Pay

For many people, relocation is all about jobs. That’s not my circumstance. I am pretty much a confirmed entrepreneur. I’m going to turn 50 next year never having held a long-term / full-time job. I think it’s safe to say that sending out resumes will be a waste of time. I’ve been self-employed for years as a small-business lawyer and a higher-ed tutor. (I’m the only known tutorney, thank you very much). Therefore, my livelihood depends on access to clients with disposable income. If rent were no concern, my ideal location would be an urban neighborhood densely packed with rich people. The nice thing about derived income is that it doesn’t matter what the local industry is. I wouldn’t have to know coding to live in San Jose, banking to live in Manhattan, or lobbying to live in DC. I’d simply offer my services to those who do. The more they earn, the more I earn — though not necessarily as much as them.

As a secondary concern, California has such high income tax that I’m starting to notice it as an avoidable drag on take-home pay.

Rent

Of course, rent is a concern. That’s why, in LA, I live as close as I can to dense, rich neighborhoods like Santa Monica and Beverly Hills without actually living in them myself. The cost of doing business takes a great bite out of my take-home pay, too; I have to pay rent for my office as well as my apartment. If income were no concern, my ideal location would be a trailer park in Mississippi. But of course, local clients would be few and far between.

That raises the interesting question of online work. I do have a couple of law-office clients now who would continue working with me remotely wherever I went. That’s not enough to keep me afloat, though, and I would have a lot of trouble finding new clients strictly online.

The goal: A lucky balance

So I need to find a location where income is randomly high and rent is randomly low. I can’t think of a place that would score well on both for obvious reasons. This will clearly require some data mining. Only after I scout out some of the luckiest locations will I know if there is a slam-dunk winner. I don’t even know whether inter-city or intra-city variation is higher yet.

Other Considerations

The sheer numbers will be only a starting point. I’d still have to figure out the logistics of getting to know the local markets and demographics, finding new business, commuting from home to work, etc. Starting business over is very difficult. I know LA pretty well, and, if nothing else, I am established here. A major relocation would involve a leap of faith. It’s also important to consider the supply side of my industries, i.e. the density of test-prep centers or attorneys nearby. I am only licensed to practice law in California, so moving to a different state would also involve the extra step of taking another bar exam.

Locales

Here are some obvious candidates for my first round of comparisons:

• Locations known for high income (Malibu, NYC)
• Locations known for low rent (Palmdale, Detroit)
• Locations toted as having a favorable “index of living” or “local purchasing power”
• Locations as specific as possible.
• Numerous neighborhoods within the LA metropolitan area and numerous locations elsewhere.
• And of course, I’ll also see how my family’s locales stack up. I have a brother, sister, and parents in three completely disparate communities in different states.

Variables

For each location, key variables will include median income per capita and median rent. Those figures are available for almost every city, county, and metropolitan area in the US at www.DeptOfNumbers.com , which uses data from the US Census Bureau’s American Community Survey (which apparently updates annually). I might also be able to find a breakdown of income and / or rent by US Census tract; income is mapped out very nicely at https://project.wnyc.org/median-income-nation/#5/38.630/-95.890 .

This will obviously be an ongoing series. Next time, I will post my initial findings. Let me know if you have any information about affordable locales or thoughts about my modeling.