Analysis: The American Household Surplus Gap (BLS Data)

The Household Surplus Lab is an interactive simulator designed to answer a critical question: After covering the absolute basics, how much money do U.S. households actually have left, and what would that capital become if it were invested?

Using data from the Bureau of Labor Statistics (BLS) spanning from 2000 to 2024—with projections extending to 2026—the tool visualizes the divide between those who can build wealth and those who are fundamentally short on essentials.

🛠️ How the Simulator Works

The tool allows users to manipulate various economic levers to see how different demographics fare. The core logic follows a simple mathematical path:

$\text{Household Surplus} = (\text{Market Income} + \text{Benefits}) - \text{Essential Expenses}$

1. Configuration Settings

Users can customize the scenario using the following parameters:

Category	Options / Controls	Details
Income Groups	Quintiles, Quartiles, Deciles, or Custom Splits	Compare the bottom 50% vs. top 10%, or the top 1% vs. the rest.
Income Sources	Market Income only $\leftrightarrow$ Full Benefits	Includes Social Security, SNAP, SSI, pensions, and veterans' benefits.
Essential Costs	Food, Housing, Transport, Healthcare, Insurance, Education	Users can edit specific expense amounts.
Childcare	Toggle On/Off	Accounts for the high variance between infant center care and after-school care.
Currency Base	Nominal $\leftrightarrow$ Real 2000 $\leftrightarrow$ CPI-U	Adjusts for inflation using a custom base year or constant rates.

2. The Investment Engine

The simulator doesn't just look at yearly cash flow; it calculates the compounded growth of that surplus.

Default Return: Based on historical S&P 500 total returns (approx. 8.0%).
Customization: Users can input their own assumed annual return.

3. Tax Estimation Logic

Taxes are handled via several methods to ensure accuracy or allow "what-if" testing:

CBO federal rates: Consistent federal income tax rates.
BLS pre-2013: Based on reported data.
BLS 2013–15: Based on the BLS tax model.
Advanced Options: Includes a "Simple minimum income tax" or a "U.S.-style AMT" (Alternative Minimum Tax).

📊 Visualizing the Results

The simulator generates three primary insights:

The Surplus Line: A dark $0 line serves as the threshold. Anyone below this line spends more on essentials than they earn.
Portfolio Growth: This shows the "wealth gap" in action—how a small yearly surplus for the wealthy transforms into a massive portfolio over decades.
Wealth Distribution: A comparison of net worth per household, ranging from the bottom 50% to the top 0.1%.

Wealth Tax Simulations

The tool includes a "What-if" module for wealth tax designs:

Flat Rate: Inspired by Warren Warren/Sanders.
Minimum Rate: Inspired by Gabriel Zucman.

Expert Debate on Revenue: Estimates for a tax of 2% over $50M and 3% over $1B vary wildly. Saez & Zucman estimate $\approx \$ 250 $billion/year, while the Penn Wharton Budget Model suggests$ $2.1\text{--}2.7$ trillion over a decade. Conversely, critics like Summers & Sarin (2019) argue actual collections would be significantly lower.

🔍 Methodology & Data Integrity

Data Sources

Real Data: The primary engine uses the BLS Consumer Expenditure Survey. This provides the raw numbers for income, household size, and spending across six core categories.
Interpolation: While quintiles are exact, finer splits (like the top 1%) are estimated using a smooth income curve.
Wealth Data: Net worth figures are sourced from the Federal Reserve.

Data Anomalies & Corrections

The tool accounts for several known "breaks" in the BLS timeline:

Pre-2004: Only households that fully reported income were included.
Post-2004: BLS began filling in missing income data.
2013 Shift: BLS switched to a tax model rather than asking households directly, causing a visible "dip" in after-tax income for top earners. (The simulator fixes this by using CBO rates by default).
2024 Gap: BLS did not update its tax model for 2024; the tool applies 2023 rates to 2024 data.

Future Projections (2025–2026)

Since current data ends in 2024, the simulator projects forward by:

Applying CPI-U (Consumer Price Index) to all incomes and costs.
Using 11 months of price data for 2025.
Using Jan–Aug data for 2026.
Applying YTD S&P 500 returns (through Sept 14) for 2026.

⚠️ Known Limitations

It is important to note that ~~BLS data is perfectly accurate~~. In reality, income—particularly at the bottom of the spectrum—is likely understated.

The Discrepancy: The BLS survey often shows lower-income households spending more than they earn. This suggests unreported income, debt, or family assistance. For comparison:

BLS (2024): Lowest 20% earn $\approx \$ 16,700$ (pre-tax).
CBO (2022): Lowest 20% earn $\approx \$ 26,200 $(pre-transfers) and$ \approx $44,800$ (after transfers).

Note: The CBO figures are higher because they include in-kind benefits such as Medicaid, CHIP, and SNAP.

Disclaimer: This tool is an illustrative model and does not constitute professional tax or investment advice.