Wealthier places give more in raw dollars, so any honest comparison has to hold income constant. This analysis does that, using the tax returns of entire neighborhoods. In America's identifiably Jewish neighborhoods, households report giving away roughly four times as much of their income as demographically similar neighborhoods next door.
The finding comes from a deliberately modest source: the Internal Revenue Service publishes totals of what tax filers report, summarized by ZIP code. No names, no religion, no individual returns. Just the arithmetic of a place. When you line up 29 Jewish-community ZIP codes against carefully chosen twins with the same income profile, the gap in reported charitable giving is large, consistent, and grows with income.
Matched on income. Not matched on giving.
If two neighborhoods earn the same, do they give the same?
Each Jewish-community ZIP was paired with three same-state ZIP codes chosen to look like it on paper: the same number of tax returns, the same average income, the same spread across income brackets. The paired neighborhoods earn almost exactly the same. What they report giving could hardly be more different.
Two panels below, four bars, and the same two groups of neighborhoods in each. In both panels the teal bar on the left is the Jewish-community ZIP codes and the tan bar on the right is their matched controls. The left panel measures average income per tax return on a scale to $180,000. The right panel measures charitable giving as a share of that income on a scale to 5%. In both, taller means more.
Compare the two bars in the left panel, then the two in the right.
The same comparison, in dollars per tax return. The first two boxes are income, treated then control. The last two are charitable giving, treated then control.
The two groups sit within a few percent of each other on income. On giving, they are in different worlds. And this is not a story of a handful of large donors: in the Jewish-community ZIPs a larger share of returns reports any charitable gift at all, in every single income bracket.
Seven of the ten sit in one metro area
Is this one unusual place, or a pattern across the map?
The 29 treated ZIP codes cluster into ten named communities across five states. Most sit in the New York and New Jersey suburbs; the rest are in Chicago, Baltimore, and Los Angeles.
On the map each circle is one community. Its size is the total number of charitable dollars reported there, so a bigger circle means more money. Its colour is the giving rate, the share of income given away, running from pale sand at about 1.5% up to deep teal above 10%. Size and colour are two different measurements and they do not have to agree. The dashed gold rectangle over the New York and New Jersey coast marks the area redrawn at larger scale in the panel to its right.
Find the darkest circles first, then check how big they are.
The gap widens as income rises
Does the gap hold at every income level?
Break the comparison down by income bracket and a pattern emerges. Among lower earners the two groups give at similar, modest rates. The higher the income, the wider the gap opens. Among the highest earners, those reporting $200,000 or more, Jewish-community filers give 7.15% of income to charity against 1.80% in the controls, a full four-to-one difference on its own.
The chart below splits both groups into the six income bands the IRS publishes. Each band gets a pair of bars: teal on the left for the Jewish-community ZIPs, tan on the right for their matched controls. Height is charitable giving as a share of income inside that band, on a scale to 8%. The gold figure printed above each pair is the ratio between the two bars, how many times more of its income the teal side gives.
Read the gold numbers from left to right.
Because giving climbs so steeply with income, the dollars pile up at the top. The highest bracket alone accounts for roughly 87% of all reported charitable contributions in these neighborhoods, a reminder that the tax-return measure captures the giving of itemizers, who skew toward higher incomes.
The chart below is a single bar holding every charitable dollar reported in the Jewish-community ZIP codes. It is cut into six segments, one per income band, lowest earners at the left and highest at the right, each segment darker than the one before. The width of a segment is that band's share of all the giving. Only the two widest carry their own labels.
Notice how little of the bar is left over for the first four bands.
The most generous places aren't the richest
If the gap grows with income, is this just about having money to spare?
If giving were only about having money to spare, the wealthiest communities would give the largest share. They don't. Lakewood and Monsey, both intensely observant communities with only moderate average incomes, report giving more than 10% of income, the highest rates in the sample. The wealthiest communities here, in Northern Bergen and Los Angeles's Westside, give among the smallest shares. Capacity matters, and it isn't the whole story.
Each circle in the chart below is one of the ten communities. Left to right is average income per tax return, so further right is a richer neighborhood. Up and down is the share of income given away, so higher is more generous. A circle grows with the total dollars that community reports giving, and it is coloured on the same pale-to-deep-teal scale as the map. If money alone drove generosity, the circles would run in a line from the bottom left corner to the top right.
Follow the top edge of the chart from left to right.
The same ten communities appear again below, this time as one bar each, sorted from the most generous at the top to the least at the bottom. Bar length is giving as a share of income on a scale to 11%, the colour is the scale used on the map and the scatter, so a longer bar is also a darker one, and the figure at the end of each bar is the rate.
Read the names down the left-hand side.
Arranged twins, and what they can't control for
How do you compare two neighborhoods fairly?
The challenge with comparing neighborhoods is that they differ in a hundred ways at once. A place might look generous simply because it is rich. The technique here, called propensity-score matching, is a way of comparing like with like. Think of it as arranged twins.
Start with what tax filers report
The IRS publishes, for every ZIP code, how many returns were filed, the total income, and the total charitable contributions claimed on itemized returns (Schedule A). We add up the six income bands into one figure per ZIP.
Name the Jewish-community ZIPs
Twenty-nine ZIP codes across ten well-known Jewish neighborhoods, from Borough Park to Lakewood to Pico-Robertson, form the "treated" group. These are geographic proxies, not a census of Jewish households.
Find each one's look-alikes
From every other ZIP in the same five states, the method finds three that most closely match each treated ZIP on the number of returns, average income, and income mix, without ever looking at giving. Income, not generosity, drives the pairing.
Compare the giving
Only now do we compare charitable contributions between the matched groups. Because the neighborhoods were made alike on income first, the remaining gap in giving cannot be explained away as one group simply being richer.
What this does and doesn't say
- It's an association, not a cause. Matching removes differences in income and size, but not everything. Household composition, age, religious practice, and local institutions still differ and are not measured here.
- It measures reported, itemized giving. Schedule A contributions miss cash given by non-itemizers, donor-advised funds, in-kind gifts, and tuition. The true totals differ; the comparison between matched places is what's meaningful.
- ZIPs are places, not people. A treated ZIP includes non-Jewish residents too. This is a neighborhood-level pattern, not a claim about any individual household or about Jewish identity as a cause.
- The exploratory second look is tentative. A separate comparison of Detroit and Cleveland finds a smaller positive gap (about half a percentage point), but rests on just two communities and should be read with caution.
So: do some neighborhoods give more, or are they just richer?
They give more, and the income is matched. What the tax data can show, cleanly and reproducibly, is that reported charitable giving is concentrated in a small number of community geographies far beyond what their incomes alone would predict. It is a portrait of where generosity, as the tax code records it, is unusually dense.
