Kansas City Ranks Second-Worst in Budget Efficiency Among Peer Cities, Machine Learning Analysis Finds

Analysis Date: December 2025
Methods: K-Means Clustering, Isolation Forest, PCA, Multivariate Regression, Monte Carlo Simulation, Benford’s Law

Key Findings

Inefficiency Index: Kansas City ranks 2nd worst out of 12 comparable cities on a composite budget efficiency index — only Baltimore scores worse. The gap between Kansas City and the next-worst city is over 2× (70.9 vs. 32.3), indicating structural inefficiencies 111.
Per Capita Spending: Total per-capita spending of $4,921 is roughly 70% higher than the peer city average of about $2,900. Kansas City’s FY2025–2026 budget of $2.5 billion for about 508,000 residents is materially higher per resident than Indianapolis, Milwaukee, and several other peers with similar or larger populations 1–31–31–3.
TIF Subsidies: Tax-increment financing (TIF) subsidies of $24 per capita are nearly triple the peer average of about $9 per capita, driven in part by ongoing support for the Power & Light District and other projects 4–64–64–6. In 10,000 Monte Carlo simulations, Kansas City’s TIF spending per capita exceeded 100% of simulated peer outcomes.
Lawsuit Settlements: Legal settlement costs of $36 per capita (around $18 million total in the most recent year) are about 3× the peer average of $12 per capita. A recent year in which the police department initially projected $3.5 million but actually incurred $18.1 million in settlements illustrates the scale of volatility and underestimation 7,87, 87,8.
Police Overspending: K-means clustering groups Kansas City only with Baltimore in a “high spending” cluster. After controlling for crime, population, income, and poverty in a regression model, Kansas City still spends about $50 million (17%) more on police than predicted 1,8,9.
Tax Burden vs. Services: A modeled median family of three earning $75,000 faces a combined state and local tax burden of $7,875 (10.5% of income). The peer city average is roughly $7,125 (9.5%), meaning Kansas City residents pay about 11% more — even though the city spends roughly 70% more than peers 10–12.
Savings Potential: Aligning Kansas City with peer averages for TIF subsidies and lawsuit settlements alone yields a conservative estimate of $19.4 million in annual savings, or roughly $97 million over five years, without cutting core services 1,4–8.

Methods: Six Techniques to Benchmark Efficiency

To understand how Kansas City’s record budget compares to similar municipalities, I compiled data from 12 mid-sized U.S. cities with populations roughly between 500,000 and 900,000 and diversified economies:

Kansas City, Indianapolis, Columbus, Denver, Nashville, Louisville, Memphis, Baltimore, Milwaukee, Oklahoma City, Tucson, and Fresno.

For each city, I used the latest available adopted or proposed budgets, tax structures, and selected financial indicators. The core analytical techniques were:

K-Means Clustering
Grouped cities according to spending patterns (total per capita, police per capita, TIF spending, lawsuit costs, debt ratios) to reveal natural “high-spend” vs. “lean” clusters.
Isolation Forest (Anomaly Detection)
Identified cities whose spending metrics were statistically unusual relative to the group, highlighting outliers like Kansas City and Baltimore.
Principal Component Analysis (PCA)
Reduced the dimensionality of the dataset to extract the main drivers of budget variation — e.g., the relative importance of public safety versus development incentives.
Multivariate Regression
Estimated expected police spending given violent crime rates, population size, median household income, and poverty rates. The difference between predicted and actual spending (residual) measures over- or under-spend.
Monte Carlo Simulation
Simulated 10,000 random peer scenarios using approximate normal distributions for per-capita metrics. This tested how likely it is that Kansas City’s values could occur by chance.
Benford’s Law Analysis
Applied to 276 budget line items to assess whether leading digits follow expected logarithmic distributions. Significant deviations may signal data manipulation; moderate deviations often reflect rounding or policy-driven targeting.

The goal is not to replace policy judgment, but to provide an empirical, transparent baseline: how far is Kansas City from “normal” when compared to similar cities?

Finding 1: Kansas City and Baltimore Form a “High Spending” Outlier Cluster

K-means clustering of the 12-city dataset produced three natural groups:

Cluster 0 — “Efficient”:
Oklahoma City, Tucson, Fresno
Low total spending per capita
Modest subsidies and lower legal payouts
Cluster 1 — “Moderate”:
Indianapolis, Columbus, Denver, Nashville, Louisville, Memphis, Milwaukee
Mid-range per-capita spending
More balanced mix of operating and capital costs
Cluster 2 — “High Spending”:
Kansas City and Baltimore
Elevated total spending per capita
High police costs
Higher-than-average subsidies and liabilities

The most notable result is that Kansas City clusters only with Baltimore, which has well-documented fiscal challenges and high legacy costs 2,3,132, 3, 132,3,13. Kansas City does not cluster with geographically similar and economically comparable peers like Indianapolis or Columbus. That signals that the structure — not just the size — of Kansas City’s budget is unusual.

Finding 2: Per Capita Spending ~70% Above Peer Average

Normalizing budget size to population provides the cleanest comparison. Kansas City’s $2.5 billion FY2025–2026 budget for roughly 508,000 residents works out to $4,921 per capita 111. Across the remaining 11 cities, average per-capita spending is roughly $2,898, placing Kansas City at a 70% premium.

Key Per-Capita Metrics

Kansas City’s per-capita spending exceeds many larger or similarly sized cities. For example, budget documents for Indianapolis and Milwaukee indicate per-resident spending that is substantially lower than Kansas City’s, even after accounting for their respective service portfolios and capital plans 2,3.

Police spending is similarly elevated. Kansas City devotes $675 per resident to policing, about 45% above the peer average of $466 per resident. Only Baltimore is in a comparable range when controlling for population 1,3,9.

TIF subsidies are especially striking: Kansas City spends approximately $24 per resident on TIF-related subsidies, compared to $9 per resident among peers. The largest and most visible example is the Power & Light District, where the city issued $295 million in bonds in 2006. Revenues have consistently fallen below projections, requiring general fund subsidies totaling at least $167.1 million through April 2022, with annual payments of roughly $10–12 million expected through 2040 4–64–64–6.

Lawsuit settlements — largely tied to police misconduct cases — add another outlying cost. In a recent year, Kansas City paid $18.1 million in settlements and judgments, equivalent to $36 per resident, roughly 3× the average for peer cities 7,8.

These per-capita metrics, taken together, indicate not just elevated costs but persistent structural overspend relative to comparable cities.

Finding 3: Monte Carlo Simulation Shows Kansas City Exceeds 90–100% of Peer Scenarios

To test whether Kansas City’s extremes could be explained by random variation, I modeled peer spending patterns using Monte Carlo simulation:

Fit approximate distributions for each metric based on the 11 peer cities (mean and spread).
Simulate 10,000 peer-city scenarios (e.g., total per-capita spending, TIF per capita, lawsuit per capita).
Compare Kansas City’s actual values to the simulated distributions.

Results:

Total per-capita spending: Kansas City’s value exceeded about 91% of simulated peer scenarios.
Police spending per capita: Exceeded about 88% of simulations.
TIF subsidies per capita: Exceeded 100% of simulations; Kansas City’s $24 per capita was never matched or exceeded in any iteration.
Lawsuit settlements per capita: Also exceeded 100% of simulations; Kansas City’s $36 per capita was never matched or exceeded in any run.

When a metric exceeds 100% of simulated peer outcomes, it implies that under reasonable assumptions about how peers behave, Kansas City’s value is not just high — it is statistically extreme. The odds that all these anomalies arise purely by chance are effectively negligible. That points toward policy and structural decisions, not random noise.

Finding 4: Police Spending Is High Even After Controlling for Crime and Socioeconomics

Kansas City’s high violent crime rate is real and serious. Higher crime can justify higher police spending. To determine whether this fully explains the budget, I used multivariate regression to estimate expected police spending based on:

Population
Violent crime rate
Median household income
Poverty rate

This model explains a little over half the variation (R² ≈ 0.56) in police spending across the 12 cities. Violent crime is positively and significantly associated with higher police budgets, as expected 8,98, 98,9.

Yet even after incorporating these drivers, Kansas City still spends about $50 million more on police than the model predicts, corresponding to roughly 17% overspend relative to its expected value. That overshoot appears to be driven not by need alone, but by institutional constraints unique to Kansas City.

The biggest factor is Missouri’s Amendment 4, which effectively requires Kansas City to allocate at least 25% of its general revenue to the police department, unlike any other city in the dataset 9,149, 149,14. In practice, this translates to hundreds of millions of dollars earmarked annually for KCPD, regardless of evolving needs or alternative safety strategies. The police department is also controlled by a state-appointed board, which limits local oversight.

In contrast:

Indianapolis devotes a smaller percentage of its budget to public safety without such a rigid mandate.
Denver and Nashville, despite significant public safety needs, retain more local flexibility in how funds are split between policing, prevention programs, housing, and social services 2,11,12,

High spending has not translated to proportionally better outcomes: Kansas City’s violent crime remains elevated, and key performance measures trail peers, suggesting diminishing returns on each additional police dollar. This mismatch between spending and outcomes is a core part of the inefficiency story.

Finding 5: Tax Burden Is Above Average, Yet Services Don’t Scale with Spending

Residents naturally ask: “If the city spends more, does that just mean taxes are higher?” And, critically, “Are we getting what we’re paying for?”

Kansas City’s Unique Tax Structure

Kansas City relies heavily on a 1% earnings tax on all income earned in the city, regardless of whether the worker resides there. This tax generates roughly $292 million per year and accounts for about 45% of the city’s general fund revenues 10,1510, 1510,15. Voters have repeatedly approved renewing this tax, signaling a willingness to fund services.

Few peer cities rely on a similar structure:

Columbus, OH has a higher municipal income tax rate of 2.5%, but residents can credit tax paid to other municipalities, which softens double taxation.
Louisville, KY levies a local income tax (around 2.2% for residents and 1.45% for non-residents), but with different structures and enforcement mechanisms.
Denver, CO does not levy a standard city income tax but imposes a small occupational privilege tax of about $5.75 per month (~$69 per year) on workers 111111.
Cities such as Nashville, Memphis, Oklahoma City, Tucson, and Fresno have no local income tax at all 11,1211, 1211,12.

On the consumption side, Kansas City’s combined state and local sales tax is around 9.1%, high but not unique; Nashville and Memphis both reach around 9.75%. Kansas City’s effective property tax rate (~1.19% of assessed value) is moderate — higher than Denver or Nashville, lower than Baltimore or Milwaukee 10,1110, 1110,11.

Modeled Tax Burden for a Median Family

To combine these elements, I modeled a family of three with:

$75,000 annual income
A median-value owner-occupied home
Typical spending on taxable goods and services

For each city, the model sums expected state and local taxes: income or earnings tax, property tax, sales tax, and relevant local levies.

Kansas City: ~$7,875 total taxes (10.5% of income)
Peer City Average: ~$7,125 total taxes (9.5% of income)

In this model, Kansas City ranks 4th highest out of the 12 cities in tax burden as a share of income. Cities like Baltimore, Columbus, and Milwaukee are somewhat higher for structural reasons (e.g., property tax rates and state income tax), while cities like Tucson and Fresno are substantially lower 10–1210–1210–12.

The Tax-to-Service Efficiency Gap

The key problem is not just that Kansas City’s tax burden is slightly above average — it’s that spending is vastly above average:

Taxes: ~11% higher than peer average
Spending: ~70% higher than peer average

This implies that every dollar residents contribute delivers less service than it should compared to peer cities. Several structural factors explain this gap:

Subsidy Leakage
Excessive use of TIF and other developer incentives shunts tax dollars into underwriting private projects that do not always deliver promised public benefits. The Power & Light District alone has required more than $167 million in general fund subsidies over time and continues to demand around $10–12 million per year 4–64–64–6.
Lawsuit Liabilities
High police misconduct settlements — like the recent $18.1 million payout year — act as a recurring drain. Every spike in settlements crowds out basic services such as streets, housing, and parks 7,87, 87,8.
Mandated Police Spending
The state-imposed 25% police funding floor locks a large share of the budget into a single department 9,149, 149,14. This rigid allocation is not sensitive to changes in crime patterns, growth priorities, or the emergence of non-police interventions (outreach workers, mental health response, etc.).
Debt and Legacy Costs
Historical borrowing for projects that underperformed (like certain downtown developments) increases annual debt service. Dollars used for interest and principal repayments do not show up as new, visible services, but they still consume tax revenue 1,4,51, 4, 51,4,5.

The result is an efficiency gap: residents are paying near-top-tier tax rates while the city’s cost structure, incentives, and liabilities prevent these revenues from translating into proportional public goods.

Composite Efficiency Index: Kansas City Ranks Second-Worst

To summarize the picture across multiple dimensions, I created a composite efficiency index combining:

Spending efficiency (per-capita and relative to peers)
Police spending efficiency (actual vs. model-predicted)
TIF burden (per-capita subsidies vs. peers)
Lawsuit burden (per-capita settlements vs. peers)
Debt ratios
Pension funding status and legacy obligations

Each factor was normalized and weighted according to addressability (how much policy can realistically change it) and fiscal impact. Higher scores indicate worse performance.

Ranked Scores (Higher = Less Efficient)

Baltimore — 100.0
Kansas City — 70.9
Milwaukee — 32.3
Fresno — 27.7
Denver — 26.9
6–12. Others below 20.0

The gap between Kansas City (70.9) and Milwaukee (32.3) is stark — more than double. This suggests Kansas City’s issues are significantly deeper than a “top three” label might imply. It is not just on the wrong side of the median; it is much closer to Baltimore than to any other peer on this composite measure 1–3,10–131–3, 10–131–3,10–13.

Benford’s Law: No Obvious Signs of Fraud, But Deviation Reflects Rounding and Targets

To test for potential manipulation in the underlying budget numbers, I applied Benford’s Law, which predicts the distribution of leading digits in naturally occurring datasets. While not definitive, it is a widely used forensic tool in accounting and tax enforcement.

Results:

Sample Size: 276 budget line items
Chi-square statistic: 35.04
P-value: < 0.05 (indicating statistically significant deviation)
Mean Absolute Deviation: 3.28%

In general guidance, a mean absolute deviation between 2% and 5% is considered “acceptable”, while values above 5% may warrant closer scrutiny. Kansas City’s 3.28% falls within the acceptable range, suggesting that deviations likely result from rounding and policy-based targets (e.g., round-number allocations or “25% of revenues” rules) rather than overt fabrication 161616.

In other words: the budget is inefficient, not necessarily fraudulent. The structural problems are policy choices, not cooked books.

Savings Potential and Reform Opportunities

If Kansas City simply brought its most extreme anomalies back in line with peer averages, the city could realize substantial savings without cutting core services.

Estimated Annual Savings

TIF Subsidy Reduction:
Bringing per-capita TIF subsidies closer to the peer average could save approximately $7.5 million annually (with uncertainty depending on contract terms and renegotiation timelines).
Lawsuit Settlement Reduction:
Reducing per-capita settlement costs to a peer-like level could save approximately $11.9 million per year, based on historic volatility and comparable cities’ experiences.

Taken together, a conservative combined estimate is $19.4 million in annual savings, or roughly $97 million over five years. These savings do not require cutting basic services; they come from stopping leakage into underperforming subsidies and avoidable legal liabilities 1,4–81, 4–81,4–8.

Policy Recommendations

Independent TIF Audits

Audit all active TIF districts for performance against original projections.
Implement mandatory sunset or restructuring for chronically underperforming projects.
Tighten “but-for” criteria so only projects that truly need subsidy qualify.

2. Strengthen Police Oversight and Risk Management

Build an actuarially sound settlement reserve in the budget to prevent surprise shortfalls.
Invest in training, accountability, and early warning systems to reduce the underlying incidents producing lawsuits.
Use the city’s budget leverage and public pressure to push for governance reforms under the state-controlled police board.

3. Peer Benchmarking and Transparency

Publish an annual “Kansas City vs. Peer Cities” report showing per-capita spending by category.
Use this report to inform budget debates and prioritize reforms with the greatest potential savings.

4. Improve Priority-Based Budgeting (PBB)

Integrate explicit efficiency metrics (cost per outcome, not just cost per program).
Require departments to demonstrate both impact and efficiency to maintain or expand funding.

4. Link Savings to Tax Relief or Visible Investments

Earmark a portion of TIF and lawsuit savings for targeted property-tax relief or earnings-tax mitigation for low-income workers.
Alternatively, commit to visible, high-impact investments (e.g., street repair, neighborhood infrastructure) so residents see direct benefits from improved efficiency.

Conclusion

Kansas City’s fiscal story, viewed through data and machine learning, is clear: it behaves like an outlier among its peers. Residents pay more than their counterparts in similar cities, yet the city spends dramatically more than its peers and still struggles to deliver commensurate value.

It clusters with Baltimore, not with Indianapolis, Columbus, or Denver.
It repeatedly exceeds 90–100% of simulated peer scenarios for key spending metrics.
It devotes outsized shares of its budget to mandated police funding, subsidies, and legal settlements.

The good news is that these are human-made constraints, not immutable laws of nature. Kansas City can:

Renegotiate or reform underperforming development incentives.
Reduce lawsuit exposure through better oversight and policy.
Push for governance reforms that restore local control over a quarter of its general fund.
Use data-driven benchmarking to align with proven best practices from other cities.

Machine learning doesn’t make policy — but it can highlight where resources are being misused and how far a city has drifted from reasonable norms. In Kansas City’s case, the drift is substantial, but the path back to efficiency is visible: tackle TIF, tackle lawsuits, rethink mandates, and commit to transparent, comparative budgeting.

For residents, the takeaway is simple and powerful: there is real money on the table. If Kansas City can close even part of the gap between what it spends and what peers spend, while maintaining services, it can lower pressure on taxpayers or finally fund the investments that residents say they want most. The numbers do not guarantee change — but they make a compelling case that change is overdue.

About the Author

Brian Curry is a Kansas City–based AI Strategist, Marketing Economist, and Applied Researcher. He is the founder of Vector1 Research, an independent lab advancing marketing economics, causal inference, cognitive AI systems, and the economics of intelligent automation.

He has led data science and analytics initiatives across major enterprises including Koch Industries ($125B), Tractor Supply ($14B), Vail Resorts, Hallmark, Garmin, and AT&T. His work applies economic thinking and causal inference to connect marketing investment to outcomes, operational decisions to economic value, and AI adoption to competitive advantage.

Research contributions:

Papilon — Open-source platform for marketing analytics, causal inference, and simulation

Memory-Node Encapsulation (MNE) — Cognitive AI architecture for artificial episodic memory

The Autonomy Economy — Macroeconomic framework for AI-driven transformation

Daedalus — Platform for causal and economic analysis

Contact: brian at vector1.ai | Vector1 Research | LinkedIn

References

1 City of Kansas City, Missouri — FY2025–2026 Adopted Budget and Financial Plan.

2 City of Indianapolis / Marion County — 2024 Budget Book and public safety appropriations.

3 City of Milwaukee — 2024 Executive Budget Summary and per-capita spending analysis.

4 City of Kansas City, Missouri — Power & Light District financing and TIF bond disclosures.

5 Missouri State Auditor and related reports on Kansas City tax-increment financing performance.

6 Brookings Institution — Analysis of Missouri’s TIF program and “but-for” criteria in development incentives.

7 Public reporting and council briefings on Kansas City Police Department lawsuit settlements, including cases where projected settlement needs were significantly understated relative to actual payouts.

8 Local investigative journalism on Kansas City police misconduct, civil settlements, and their fiscal impact.

9 State of Missouri — Amendment 4 ballot language, legislative history, and guidance on the 25% police funding mandate for Kansas City.

10 Tax Foundation — State and local tax burden comparisons for U.S. cities and states.

11 Denver, Nashville, Memphis, and related city budget and tax documents outlining sales, property, and local occupational privilege taxes.

12 District of Columbia Office of the Chief Financial Officer — Comparative Tax Burden Studies for representative U.S. cities.

13 City of Baltimore — Budget reports and audits detailing fiscal challenges, spending patterns, and legacy costs.

14 Missouri and Kansas City legal and policy analyses of the governance structure of the Kansas City Police Department and the implications for local budgeting.

15 Kansas City revenue reports and ballot summaries describing the earnings tax, its share of the general fund, and voter approval history.

16 Forensic accounting and audit literature explaining Benford’s Law thresholds and interpretation of mean absolute deviation ranges for financial datasets.