About This Tool

This tool is the product of a collaboration between the Cornell ILR Ithaca Co-Lab, Cornell ILR Buffalo Co-Lab, and the Tompkins County Living Wage Coalition.

Generally speaking, a living wage is the minimum hourly amount that a full-time (2,080 hours/year) worker must earn to afford basic necessities in their geography of residence, without the need for public or private assistance. Whereas different researchers estimate living wages using different techniques, one common thread that runs through most living wage calculations is that they rely on consumer expenditure data.

For example, arguably the most well-known and widely-used set of living wage estimates from the MIT Living Wage Calculator are generated with spending data that are published in national publicly-accessible annual and semiannual surveys. Drawing on data for nine categories of “basic needs” spending – food, childcare, healthcare, housing, transportation, civic engagement, broadband Internet, miscellaneous items, and taxes – the MIT Calculator adds up the annual amount a typical household would need to cover the costs of these items. The resulting sum represents a “basic needs budget” for a given household. Dividing that annual amount by 2,080 hours, or the approximate number of hours worked by a full-time employee in a calendar year, the MIT Calculator reports, for each county in the United States, the living wage associated with a basic needs budget. Rather than reporting a single dollar figure, however, the MIT Calculator models basic needs budgets for a variety of household scenarios that depend on the number of adults, working adults, and children living in a household. Thus, the MIT Calculator does not report a single living wage for each county; but a schedule of living wages for a given county that illustrates how living wages vary for different household circumstances.

The point of highlighting the MIT Calculator here is twofold. First, it has come to be recognized as something of a gold standard in the U.S. living wage discourse, thereby making it a relevant benchmark against which to compare other living wage estimates. Second, its “basic needs budget” methodology is generalizable and is frequently engaged in other living wage estimation projects.

One such project, which pre-dates the MIT Calculator by nearly a decade, is the annual living wage calculation prepared for Tompkins County, NY. Since 1994, a Tompkins County living wage coalition has drawn on state- and national-level survey data – in most cases, the same data sources that power the MIT Calculator – to estimate a living wage for a single adult living alone in Tompkins County. Similar to the MIT Calculator, the Tompkins County Living Wage is based on data for nine basic needs spending categories: housing (rent), food, transportation, communication, healthcare, recreation, savings, miscellaneous items (e.g., housekeeping supplies, apparel, etc.), and income taxes.

For nearly thirty years, estimation of the Tompkins County Living Wage was led by Alternatives Federal Credit Union (AFCU) in Ithaca, NY. Following AFCU’s 2021 living wage report, the Tompkins County Workers’ Center took over the estimation project for 2022. In all cases, the methodology remained consistent over the years. Similar to the MIT Calculator, the Tompkins County coalition collected annual household spending data for the nine categories of basic needs listed above, summed up the expenses to generate a basic needs budget, and divided by 2,080 hours to compute a living wage estimate. Data on these spending patterns came largely from national, publicly available datasets published by the Bureau of Labor Statistics (BLS).

Whereas the same BLS datasets used for the Tompkins County living wage project also feature prominently in the MIT Calculator, and are therefore both highly relevant and have established utility for creating living wage estimates, they are also characterized by a relatively coarse spatial resolution. Specifically, BLS consumer spending data are essentially only available at the national- or statewide- scales. In other words, BLS spending data do not permit researchers to localize living wage estimates down to the county level. Consequently, drawing on these data to construct county-level living wage estimates in New York State means that all counties will exhibit identical annual budgets in several spending categories, regardless of where the counties are situated. For instance, current MIT Calculator estimates show that in both Tompkins County and New York County (i.e., Manhattan), annual costs of food, medical care, transportation, civic participation, and miscellaneous items are exactly equal, despite the two counties’ vastly different spatial contexts. The only cost differences appear in housing and taxes, for which location-specific data are available from non-BLS data sources.

Importantly, the above is not meant as a critique of the MIT Calculator – which, again, is arguably the gold standard of living wage estimates – nor is it a critique of past calculations carried out by the Tompkins County living wage coalition. Rather, it is merely an observation that what are, on balance, the best publicly-accessible datasets available to compute living wages (namely, BLS consumer expenditure data) can hamper efforts to generate regularly-updatable localized wage estimates due to their coarse spatial resolutions. For that reason, the Cornell University-based research team overseeing the current (2023) iteration of the Tompkins County living wage calculation is proposing a slightly new strategy for 2023 and beyond. As detailed in the next section, that strategy is grounded heavily in a premium, commercially-owned consumer expenditure dataset to which Cornell purchases annual access.

Cornell ILR Methodology for 2023

In the interest of maintaining as much tradition as possible, the proposed strategy for estimating the 2023 living wage for Tompkins County starts with generating a basic needs budget from the nine spending categories that have been used to perform this task ever since AFCU generated its first countywide living wage estimate in 1994. As noted in the previous section, those spending categories are:

1. housing (rent),
2. food,
3. transportation,
4. communication,
5. healthcare,
6. recreation,
7. savings,
8. miscellaneous items (e.g., housekeeping supplies, apparel, etc.),
9. and income taxes.

Basic Needs Budget Data

Of the above-listed nine categories, housing (#1) and taxes (#2) are the only two data points that are fully localized and based on information specific to Tompkins County. With respect to housing, the fair market rent (FMR) for a one-bedroom apartment – which is the housing expense that AFCU and the Tompkins County Workers’ Center have used to estimate an annual basic needs housing budget for almost three decades – is available from and annually updated by the U.S. Department of Housing and Urban Development (HUD) for all counties in the United States. Concerning taxes, because Tompkins County does not collect local income taxes, the federal and state income tax bills for a living wage worker in Tompkins County is straightforward to compute using publicly-available information on federal and state marginal income tax rates.

Unlike housing and income taxes, data for spending categories #2 through #8 in Tompkins County have historically come from the national BLS Consumer Expenditure Survey (CE), with supplemental local information sprinkled in whenever possible. Recall, though, that the CE data are not capable of being disaggregated and tuned specifically to Tompkins County. For that reason, the authors of this report propose to use frequently-updated, fine-resolution consumer expenditure data from a commercial data vendor called SimplyAnalytics. SimplyAnalytics is a web-based interactive mapping portal and data provider that allows users to visualize, summarize, and otherwise access data from a variety of public (e.g., U.S. Census Bureau) and private (e.g., Experian Simmons) sources. The private sources include, among other things, data from market research firms on consumer spending patterns. Because these data are collected for firms seeking to better understand geographic patterns of consumer buying power and make critical investment decisions based on those patterns, the data are made available at relatively fine spatial resolutions. Whereas most such datasets are available down to the census block group level of analysis, for present purposes, they offer localized estimates of consumer spending at the county scale. As a result, the SimplyAnalytics data can plausibly be used to generate more localized, county-specific household budgets (and, it follows, living wage estimates).

That being said, there is a trade-off involved in using these localized spending data: to achieve the superb geographic and temporal precision offered by SimplyAnalytics, it is necessary to scale a paywall. More specifically, SimplyAnalytics is a commercial, subscription-based service, and access to the service’s consumer spending data can be costly. As such, any living wage calculation methodology that involves the use of these data can only be replicated by users who purchase access from the data vendor. Whereas numerous universities across the U.S. do maintain active SimplyAnalytics subscriptions, the tool and its data are presumably out of reach for most community-based organizations and coalitions involved in living wage estimation. Thus, to replicate or expand on the methodology used herein, it may be necessary for living wage researchers to work with academic or corporate partners who can tap into the commercial data; otherwise, such researchers would either need to purchase the data on their own, or rely on coarser- (e.g., national- or state-level) resolution public datasets from agencies like the BLS.

Basic Needs Budget Methods

Between HUD data on fair market rent (FMR) for Tompkins County and SimplyAnalytics data on county-level spending patterns with respect to the remaining “basic needs budget” categories, most information needed to construct a basic needs budget (and, it follows, a living wage) estimate for Tompkins County are readily obtainable for this project. However, there is at least one more hurdle to surmount prior to moving forward.

As the name implies, a basic needs budget is associated with a basic, or relatively low-cost, lifestyle. In practice, this observation means that most living wage calculations are not done for “typical” (e.g., average or median) spending habits, but for below-average spending habits. HUD’s FMR data, for instance, are legislatively set at the 40^th percentile of gross rents in a given location. Echoing this observation, past calculations by the Tompkins County living wage coalition have drawn on 40^th percentile (i.e., second quintile) expenditure values from the BLS CE to create annual budget estimates for many of the non-housing spending categories represented in the coalition’s basic needs budget calculation.

Whereas BLS datasets report 40^th percentile expenditure values alongside average values at national and statewide scales, localized (e.g., county-level) expenditure data from SimplyAnalytics are only reported for averages. Thus, to take advantage of the localized, county-specific data available through this commercial provider, it is necessary to estimate 40^th percentile expenditures from household averages. One way to accomplish this objective is to draw on current (2017-21 Five-Year) U.S. Census American Community Survey (ACS) data on average household income by quintile. As summarized in the top row of Table 1, below, the current ACS reports that average household income in Tompkins County, expressed in 2023 dollars, is approximately $102,969. Further, in Table B19081, the ACS reports that average income for Tompkins County households that fall in the second income quintile (i.e., 40^th percentile) is, in 2023 dollars, $38,461.

Drawing on these data points, one reasonable strategy for estimating 40^th percentile expenditures from SimplyAnalytics data is to multiply average expenditure values for Tompkins County by the observed ratio of mean income in the second quintile ($38,461) to mean overall household income ($102,969) in the county. That ratio, which equals approximately 37.35%, functions as a straightforward adjustment factor that is used in Table 1 to approximate 40^th percentile expenditure levels (column 3) from average expenditure levels (column 2) in Tompkins County. The final column of Table 1 (column 4) presents corresponding expenditure values from the 2022 living wage report for Tompkins County, which was prepared by the Tompkins County Workers’ Center.

Table 1. Estimated 40^th Percentile Monthly Expenditures in Select Basic Needs Categories for 2023, Compared to 2022 Expenditure Values

*Known value from the current U.S. Census American Community Survey; ^{^}Estimated value derived by multiplying the ratio of average income in the second quintile to average overall income by the overall average expenditure value; ^#Communication is the sum of $64.10 – the average lowest cost broadband plan across Tompkins County census tracts, as reported by Broadband Now – and the SimplyAnalytics expenditure data point for telephone services in Tompkins County; ^&The average cost of a commercial or Blue Cross/Blue Shield health plan in Tompkins County, as reported in the SimplyAnalytics dataset, is $145.76. This value was assumed to be constant across income quintiles (e.g., the NYS healthcare marketplace reports a similar value of roughly $159 per month for a Silver-level plan for an individual earning approximately $38,000). The difference between the average (observed) and 40^th percentile (estimated) expenditure values in healthcare is therefore attributable to out-of-pocket expenditures.

As should be clear from the data reported in the final two columns of Table 1, the 40^th percentile approximation method, fueled by localized expenditure data, yields expenditure levels that are both compatible with and closely mirror expenditure levels reported for 2022 at the state and national scales. Stated another way, the technique produces estimates that are at once reasonable and seemingly reliable when juxtaposed with aggregated BLS data, all while possessing the distinct advantage of using fine resolution data that are specific to Tompkins County.

Triangulation

Although the localized basic needs budgeting method described above represents the official method of this report – and the resultant wage for a single worker will be advanced as the 2023 “living wage” for Tompkins County – the research team acknowledges that there is no single, nor no optimal, way to calculate a living wage. As such, in addition to computing and presenting the 2023 Tompkins County living wage as derived through this report’s adopted methodology, the researchers will also highlight three additional living wage values for Tompkins County for context and comparison.

Alternative Measure #1: The MIT Calculator Living Wage for a Single Adult with No Children

Insofar as the MIT Living Wage Calculator was already introduced and discussed above, there is little need here for additional unpacking of what this resource is and the information it conveys. However, it is worth restating that the MIT Calculator does not report a single living wage for counties in the U.S., but a range of living wages that cover various household compositions. More specifically, for each county, the MIT Calculator reports twelve living wages for twelve scenarios that cover up to two adults and three children in a given household. Four scenarios relate to households in which there is just one, single working adult and between zero and three children. An additional four scenarios relate to households in which there are two adults, only one of whom works, and between zero and three children. And the final four scenarios relate to households in which there are two adults, both of whom work, and between 0 and three children. Because, historically and continuing with this report, the Tompkins County living wage coalition is concerned with a living wage estimate for an individual worker (presumably who lives alone and does not have children), the value from the MIT schedule for Tompkins County most relevant to this report is the living wage for a single working adult with no children.

Alternative Measure #2: The “Housing Wage” as Computed with HUD Fair Market Rent Data for a One-Bedroom Apartment

According to the National Low Income Housing Coalition (NLIHC), a housing wage is the hourly wage that a full-time worker (2,080 hours per year) must earn to afford a rental home without spending more than 30% of their gross monthly income on rent. Whereas NLIHC computes the housing wage for various locations in the U.S. using the HUD Fair Market Rent (FMR) for a two-bedroom apartment, the Tompkins County living wage coalition has historically used the FMR for a one-bedroom unit in its calculations of a local living wage. Thus, the appropriate FMR-based housing wage for use as a comparator in this report is one associated with a one-bedroom unit.

The reason for the 30% threshold in computing the housing wage is that HUD, as well as the broader affordable housing community, defines housing cost burden as a situation in which households spend more than 30% of their income on housing. To be housing cost burdened is to struggle with housing unaffordability and live in a state of precarity. Insofar as a living wage is a wage that is sufficient for workers to meet their basic needs without struggling to find supplemental financial assistance, a housing wage is therefore a type of living wage – one that ensures a worker earns enough to pay for the shelter in which they currently reside.

Computing a housing wage from HUD FMR data is straightforward and does not require any additional data on expenditures in other areas of basic needs. Rather, it requires only one input: the local HUD FMR for the unit type (e.g., one-bedroom) on which the calculation is to be based. For the purposes of this report, the housing wage for a one-bedroom unit in Tompkins County at the 2023 HUD FMR level is equal to:

Housing Wage = [(Monthly FMR_beds * 12) / 0.3] / 2080 hours

where FMR_beds is the FMR associated with the desired unit type. Once again, whereas NLIHC uses a two-bedroom unit in its housing wage calculations, herein, the research team will use FMR for a one-bedroom unit in Tompkins County.

Alternative Measure #3: The “Housing Wage” as Computed with U.S. Census Bureau Data for Single Tompkins County Renters Who Live Alone

The housing wage developed by NLIHC and discussed above is powered by HUD data on Fair Market Rents (FMRs). The HUD FMR dataset reports 40^th percentile gross rents, by unit size (i.e., number of bedrooms), for all counties in the U.S. One potential challenge with using FMR data on one-bedroom units to compute county-level housing wages is that, especially in rural parts of a state, some counties might not have many one-bedroom units on which to base the calculation.

For instance, according to current (2017-21 Five-Year) U.S. Census American Community Survey (ACS) data, there are roughly 5,515 one-bedroom, renter-occupied housing units in Tompkins County. However, the same dataset reports that there are roughly 13,909 households that contain single individuals who are living alone. Thus, by focusing only on one-bedroom units, the FMR-based measure of a housing wage could potentially fail to capture the living situations for a meaningful fraction of the target population (e.g., single workers, without children, who live alone). One strategy to overcome this challenge, and to more holistically measure the costs of housing for single renters without children or roommates, is to draw on self-reported data on the rents paid by such individuals living in Tompkins County. The premier source for such data is the current U.S. Census ACS Public Use Microdata Samples (PUMS).

Briefly, PUMS “enable[s]…users to create custom estimates and tables…that are not available through ACS pretabulated data products. The ACS PUMS files are a set of records from individual people…with disclosure protection enabled so that individuals…cannot be identified.” That is, ACS PUMS datasets contain anonymized records for individual survey respondents – the data are not aggregated. The rich person- (worker-) level information contained in PUMS records allows researchers to construct detailed pictures of worker and economic conditions for numerous locations across the United States, including for Tompkins County. Among other things, PUMS data make it possible to identify single, working adults who live alone in rental housing and do not have children. In addition to providing data on their work and housing situations, all such workers report to the Census Bureau the gross monthly rent that they pay to live in their home dwelling. Accordingly, using basic descriptive statistical techniques, it is possible to identify – in the spirit of HUD FMRs – the 40^th percentile gross rent paid by single renters who live alone in Tompkins County. Upon inflating that figure so that it can be expressed in 2023 dollars, it offers yet another comparison measure of a living, or housing, wage for Tompkins County.

Summary

Despite the initial focus on generating a 2023 living wage estimate for a single adult in Tompkins County, the research team determined that performing the calculations described above for all counties in upstate New York would allow for a more comprehensive picture of cost of living across the state. The New York City and Long Island Labor Market Regions were excluded from the mapping exercise, due to vastly different costs of living downstate versus upstate. The interactive mapping application at the top of this page therefore offers a limited look at what a “living wage” for a single adult (with no children) might be throughout NYS. The bar chart in the upper-right of the tool is population-weighted. By default, the different measures are effectively “upstate” living wage measures. Using the Labor Market Region and/or County filters at the top of the tool will update the living wage measures so that they are specific to the area(s) selected by the user. Hovering over any county on the map will generate a pop-up box that shows the four alternative living measures computed for that specific county.