The Mystery of Jewish Economics
Using statistical analysis to help understand the financial problems facing the Orthodox world
Money has always been a tough problem to solve.
Frum communities, as we're regularly reminded, face no end of financial crises. Most yeshivos and Bais Yakov schools struggle to meet their payroll each month but freely admit that they're not paying their teachers nearly as much as they deserve. For too many families, genuine poverty is an unwelcome but constant companion. And even for the middle class, keeping up with tuition and chasuna obligations is a struggle - over and above the kinds of pressures common to our non-Jewish neighbors.
On the other hand, there does seem to be an awful lot of money flying around. Luxury homes sprout like mushrooms after a rain storm. More and more families seem able to enjoy affluent and even extravagant lifestyles - for the most part, all while being genuinely thoughtful and generous towards those less fortunate.
Do we have a systemic economic problem? Is there anything we could be doing better? Those are questions we can't possibly address without better understanding the facts on the ground.
The “facts on the ground” is what this article is about. Using publicly available data from various US government sources and standard data analytics tools, I'm going to try to paint a picture of how we live, where we get our money, how we spend it, and what's left over when we're done.
Fully understanding the obvious contradiction that that picture will soon present won’t be so simple.
Understanding our data
While the job of data analytics is to describe the real world as accurately as possible, it has its limits. A statistical trend is not a prediction of what will certainly happen to any one individual. Instead, it's nothing more than an indication of the behavior one might expect to see over time - more often than not - within a large population.
In our case, we're trying to understand the financial and demographic status of many thousands of Orthodox Jews without necessarily knowing them personally or even visiting their neighborhoods. To that end, I've assembled data on the dozen zip codes in New York and New Jersey where frum Jews are most highly concentrated:
08701 Lakewood, NJ08527 Jackson, NJ08753 Toms River, NJ11204 Boro Park, NY11219 Boro Park, NY11230 Midwood, NY10952 Rockland County, NY10950 Monroe, NY10977 Spring Valley, NY (New Square)11367 Kew Gardens Hills, NY11691 Far Rockaway, NY11374 Rego Park, NY (Forest Hills)Bear in mind that the populations of those zip codes are not exclusively Jewish, so the results won't be as precise as we'd like.
The data comes from the US Census' American Community Survey (ACS) from 2019 (those happy days before COVID blew everything up). We looked at household and family income, the percentage of total households that receive food stamps, and the percentage of total households with four or more occupants.
We sought to validate the specifically "Jewish" characteristics of these zip code communities by comparing their household size numbers with the national average. Whereas larger households, on average, made up only 21% of households across all US zip codes, they represented 36% within these 12 neighborhoods. This is obviously consistent with what you'd expect to see with the larger families that are common in frum communities.
But in general, all the numbers we saw from those communities showed remarkably homogeneous characteristics. All 12 zip codes were similar enough to each other and different enough from national averages to give us confidence that we were looking at "Jewish" data.
We’ll be using data representing both households and families. Households are units where any number of people live, while families are households whose occupants are related. Census data treat the two differently, so stay alert for differences in results.
What we saw
There are noticeable and significant differences between national economic status averages and those within zip codes that are home to large numbers of frum Jews:
The average income from all sources (including government benefits) for household units living together in a single home across all 30,000+ US zip codes was $77,907, while frum zip codes enjoy $85,280 - fully 9% higher.
12.5% of homes received federal food stamp support (now referred to as the Supplemental Nutrition Assistance Program) throughout the country. But the proportion of homes in the frum zip codes was 21.5%. In fact, only food stamp reliance rates in the bottom quarter of frum populations fall even close to the national average.
36% of households in our dozen communities have four or more individuals living full-time, while the national rate is only 21%.
What do the numbers suggest?
It's clear that a greater proportion of households in our communities are earning healthy incomes than in the general population. But at the same time, a much larger segment is eligible for public poverty relief. That, on a first glance, would suggest that income disparity - meaning the statistical gap between the richest and poorest individuals - within the community is larger than normal.
But it's not.
In fact, a dataset with significant income disparity should have a higher standard deviation (SD). But the national SD for household income is 34,622, while the frum SD is only 13,133. Similarly, the distance between the low and high ends of the middle 50% of income earners is around $30,000 nationally, but only around $14,000 for Jewish populations.
All of which is a fancy way of saying that, statistically, there's a relatively short distance between our rich and our poor.
Another way of demonstrating this is through the Gini index. Gini is a common measure of inequality. In a world where all people have exactly the same income, the Gini coefficient would be zero. If all wealth was concentrated in the hands of just one person, leaving everyone else with nothing, the Gini would be one. In the real world, economies fall somewhere in between.
The official 2019 ACS Gini coefficient for the entire US was 0.481. Now, bearing in mind that my methodology was a lot less sophisticated than the one used by the Census bureau (the value I got for all zip codes was 0.39), I calculated the Gini for our zip codes to be just 0.15.
In other words, by every measure we can find, we have significantly less of an income gap than do Americans as a whole. So then why are there so many of us at both the high and low-income ends?
Adjusting our numbers
On the suggestion of friends, I reconsidered my decision to compare our frum zip codes with the rest of the US. After all, if the cost of living in the New York metropolitan area is a lot higher than most of the rest of the country, then one would expect values for any zip code in the more expensive regions to be shifted.
Selecting a group of 12 non-Jewish codes that neighbor our 12 as a control group would be difficult: many neighboring areas are unusually wealthy (think: Five Towns or The Hamptons) and could skew our results. Manually excluding the wealthy areas, however, would remove the randomness of my control group.
So instead, I selected all 536 zip codes from the 16 New York and New Jersey counties that surround our 12 Jewish zip codes and ran my code again. Adding this context changed the picture and provided at least some insight into our problem:
The average family income across the entire region was $145,024, compared with the $94,465 figure of our 12 frum codes and $90,534 for the entire country. With this added perspective we can better understand things. If anything, within the context of the New York area, frum income is lower than you'd normally expect.
Given the high income averages across the region, the overall food stamp recipient rate of only 10% (compared to 12.5% throughout the US) also makes sense. And, considering the particularly high local cost of living, the 21.5% food stamp rate in our 12 frum zip codes doesn’t sound quite so crazy.
The one point that might still be a bit confusing is the income gap. The standard deviation for family income across the larger region is a whopping $67,016. Comparing that to $41,094 for the US overall is predictable. But how are we to understand the $16,198 frum value?
I'm afraid I can't answer that question, although I’ll bet some of you will have interesting insights to add. But at least we now have a better sense of the economic environment in which we live. I hope to dig a bit deeper in future posts.