Updating the California Necessities Index

AN LAO REPORT Updating the California Necessities Index GABRIEL PETEK | LEGISLATIVE ANALYST | AUGUST 2025 SUMMARY Recent Developments Require Changes to the California Necessities Index (CNI). Due to recent changes by the Bureau of Labor Statistics (BLS), the CNI, which is based on changes in the prices of select essential goods and services and used to adjust certain payment rates, can no longer be calculated using the current formula. In this brief, we provide context on the changes and propose viable remedies that conform to the original intent of the index. Our recommendation would deviate only slightly from the existing measure while still reflecting changes in prices of the goods and services identified in the current index. BACKGROUND CNI Aims to Measure Change in Cost of Basic these statistics for the country and, in certain Needs. The CNI is used to measure changes in the spending categories, for select metropolitan areas. money required to purchase certain necessities In California, the BLS publishes data for the Los in California. In contrast to other common price Angeles, San Francisco, San Diego, and Riverside indices such as the Consumer Price Index (CPI), metropolitan areas, which are all used in calculating which calculates changes in the cost of all goods the CNI. bought by broad groups of consumers, the CNI Some Statistics Required for CNI focuses on goods and services considered Discontinued as of 2025. In early 2025, the BLS essential: food, apparel, rent, household fuels and announced that it would no longer publish statistics other utilities, and transportation. for the household fuels and other utilities category Used to Adjust Benefit Rates and for metropolitan areas (but not for the United Cost-of-Living Adjustments (COLAs). Statute States as a whole). No reason was given for this ties various benefit and cost-of-living allowances change; however, the BLS periodically reassesses to the CNI. For example, the California Department the quality of each statistic they publish. As this of Social Services uses the CNI to adjust payment category is in the CNI formula, this means that the rates for foster care providers. Similarly, COLAs CNI can no longer be calculated. for California Work Opportunity and Responsibility Discontinued Statistics Make Up Only Small to Kids and Supplemental Security Income/State Part of Index… The weights placed on each Supplementary Payment recipients, to the extent expenditure category making up the CNI are set they are provided for in the state budget, are in part in California statute, with each being chosen to based on the CNI. represent the relative importance of each category Based on Published CPI Statistics. The CNI in the overall cost of basic needs. The discontinued formula is based on statistics published by the BLS category—household fuels and other utilities— using data from surveys of consumer spending. makes up about 7 percent of the total basket of To calculate these statistics, the BLS collects data necessities as of 2024. This means that changes on a sample of goods and services bought by in the prices of other categories—such as food consumers across many urban areas. They then (26 percent of the total)—have a much larger impact aggregate these data to determine how much a on the CNI than similar changes in the price of the typical consumer spends on a certain amount discontinued category. of specific goods and services. They publish www.lao.ca.gov 1 AN LAO REPORT …But Are Historically More Volatile. While In fact, in one out of every five years, the category the weights placed on household fuels and other increased by more than 8.2 percentage points. It is utilities are small relative to other categories, prices also not uncommon for household fuels and other in that category tend to experience larger changes utilities prices to increase by only 1.4 percentage than other categories of goods and services. points—in one out of every five years, these The prices for household fuels and other utilities prices increased by less than 1.4 percentage increase by 5 percentage points in California in a points. The difference between these changes typical year, similar to rent prices and higher than is 6.8 percentage points (this is the difference food, apparel, and transportation prices. However, between an 80th and 20th percentile change across household fuels and other utilities prices are more years). In comparison, the difference in these prone to large swings. It is not uncommon for typical changes is between 3 percentage points household fuels and other utilities prices to increase and 5 percentage points for food, apparel, and by 8.2 percentage points or more, as they did in rent. The transportation category experiences more each year between 2020 and 2022. similar swings: the difference for that category is 6.6 percentage points. EVALUATING ALTERNATIVES In order to continue to adjust benefit levels, the calculate the difference between each new state will need to develop a new measurement of measure and the CNI in prior years. To score the cost of basic needs. This section lays out our well on this assessment, the average of the criteria for assessing various alternatives. differences across years should be small. It is also important that these differences be Criteria symmetric: a new measure should minimize In our search for a replacement for the CNI, the extent to which changes are consistently we first restrict our attention to options that meet higher or lower than the current CNI. Without two requirements. this, the new measure could be historically Feasibility. First, new measures must be similar to the CNI but lead to significantly calculable using publicly available data that is higher or lower benefit levels over time. published in a timely manner. Therefore, we • Fidelity to Original Concept. We assess do not consider approaches that would use each proposed measure on whether it adheres proprietary data. to the concept of the current formula. That Transparency. Second, like the current formula, is, we assess whether each option measures alternatives should provide an intuitive way to judge changes in the cost of purchasing necessities. changes in the cost of necessities. This provides • Limited Future Changes. Finally, for each transparency to benefit recipients when they assess proposal, we consider whether there are likely changes to the funds they receive. to be future changes that affect the ability Assessing Alternatives. After ensuring that the to calculate an index with the new formula. options we suggest meet the requirements above, Statutory changes to formulas dictating we judge potential alternatives on several measures: benefit amounts are potentially disruptive to recipients. While future data availability • Historical Similarity. First, the new measure cannot be fully predicted, certain statistics are should—to the extent feasible—minimize statistically more reliable than others and are quantitative differences from the old measure. thus more likely to be available in the future. To assess this, we calculate each new measure using data from prior years and 2 LEGISLATIVE ANALYST’S OFFICE AN LAO REPORT OPTIONS CONSIDERED Based on the criteria discussed above, we estimate for that category instead of a metropolitan consider the following options for replacing the CNI. area estimate. Option A: Remove Discontinued Category, Option D: Discontinue CNI, Replace With Keep Other Expenditure Weights the Same. The Other Published Series. Alternatively, the simplest option for replacing the CNI is to remove legislature could decide to retire the concept of the discontinued statistics from its calculation, while the CNI, instead setting benefit rates based on not otherwise altering the formula. other broader inflation measures calculated for Option B: Reassign Expenditure Weights to Californians. We consider the most common Most Similar Category Still Published. For this price indices specific to California: the California option, the new calculation would use the same set Consumer Price Index for Urban Consumers of statistics (excluding the discontinued statistics), (CA CPI-U) and the California Consumer Price but place additional weight on the remaining Index for Wage and Clerical Workers (CA CPI-W). category that is most like the discontinued These indices are based on changes in prices for category: transportation. consumers in the state’s four largest metropolitan areas, with the California CPI-W tracking prices Option C: Use National Estimate. This option of goods and services bought by wage earners in maintains an estimate for the cost of household those areas. utilities and other fuels by using the national ASSESSMENT OF ALTERNATIVES Historical Similarity and CA CPI-W is often large. Figure 2 on the next page shows differences between the proposed Figure 1 shows the average of the difference measures in each year since 1978. In about half of between each proposed measure and the CNI years, the CNI differed from the CA CPI-U and the across each year since 1978. Overall, small changes CA CPI-W by at least 1 percentage point. Therefore, to the calculation—as reflected in options A, B, and using the CA CPI-U or CA CPI-W to adjust payments C—are more similar to the CNI than general indices. We discuss this assessment more below. Figure 1 General Indices Differ Average Historical Differences Between Proposed Substantially From CNI. Measures that deviate from the focus on Measures and Current CNI necessities, as in the CA CPI-U and Average Historical Difference Percent of Years CA CPI-W in option D, are dissimilar Measure (Absolute Value) Higher from the CNI historically. Both the Option A: Remove 0.20 40% CA CPI-U and CA CPI-W differ discontinued category from the CNI on average by about Option B: Reassign 0.24 45 expenditure weights 1 percentage point, which is about Option C: Use national 0.19 45 25 percent of average CNI growth. estimate In Many Years, Differences Option D: CA CPI-U 0.99 28 Are Large. Further, the difference Option D: CA CPI-W 1.12 28 between the CNI and the CA CPI-U CNI = California Necessities Index; CA CPI-U = California Consumer Price Index for Urban Consumers; and CA CPI-W = California Consumer Price Index for Wage and Clerical Workers. www.lao.ca.gov 3 AN LAO REPORT Figure 2 How Would Proposed CNI Measures and Current CNI Have Differed Historically? Difference between each measure and the current CNI in each year: 1978−2024 Each bar shows the number of years that Option A differed from the Current CNI by about "x" percentage points. Option A: Option B: Option C: Remove Discontinued Category Reassign Expenditure Weights Use National Estimate 12.5 10.0 7.5 5.0 2.5 -1.0 Option D: Use CA CPI−U Option D: Use CA CPI−W CNI = California Necessities Index. would likely result in significantly different benefit Fidelity to Original Concept increases in some years relative to the CNI. Each Option Presents Trade-Offs. None of Small Tweaks to CNI Lead to Only Small the options perfectly align with the concept of the Differences. In contrast, the other options we original CNI: to measure changes in the prices of consider, which make only small tweaks to the select categories of essential goods and services in current CNI formula are historically very similar California. Adopting options A and B, which remove to the current CNI. As shown in Figure 1, on the household fuels and other utilities category average, these measures (options A, B, and C) altogether, would mean that any large changes in are about 0.2 percentage points different from the prices of these goods would not be reflected the current CNI. Further, large differences (more in the CNI. On the other hand, adopting option C, than 0.5 percentage points) are rare. Over short which uses the national estimate for changes in the time periods (three years), year-to-year differences prices of household fuels and other utilities, would average out to changes very similar to the CNI make it more likely that large changes in prices of (about 0.1 percentage points). This means that any household fuels and other utilities are reflected differences would likely partially even out within in the CNI, but may reflect changes happening at three years, leaving benefit levels at a similar level the national level, not in California. Finally, option to where they would be if the current CNI had D, which uses alternate inflation measures, does been used. away with the concept of measuring changes in the prices of necessities, instead measuring changes in the prices of all goods bought by a broad set of consumers. This approach lacks nuance. 4 LEGISLATIVE ANALYST’S OFFICE sraeY fo rebmuN -0.5 0 0.5 1.0 -1.0 -0.5 0 0.5 1.0 -1.0 -0.5 0 0.5 1.0 12.5 10.0 More than More than More than More than 1p.p. lower 1p.p. higher 1p.p. lower 1p.p. higher 7.5 in 10 of 47 years in 4 of 47 years in 12 of 47 years in 4 of 47 years 5.0 2.5 -1.0 -0.5 0 0.5 1.0 -1.0 -0.5 0 0.5 1.0 sraeY fo rebmuN AN LAO REPORT For example, changes in the price of luxury cars Small Category Statistics Could Be factor in to changes in these general price indices Unreliable, Even at National Level. While it is but are not often considered a necessity. As easier for the BLS to estimate price changes for such, Options A, B, and C likely are reasonable specific categories of goods at the national level, candidates, but present different trade-offs. statistics for certain categories of goods and services could also be difficult to calculate reliably Future Changes at the national level. In particular, the fact that the Metropolitan Area Statistics Could Face BLS has discontinued publishing statistics for the Challenges. The BLS data collection process, household fuels and other utilities category for which requires obtaining prices for a sample of metropolitan areas shows that they have found goods bought by consumers across the country, it difficult to collect a sufficient sample of prices makes it difficult to calculate reliable estimates in this category. It is possible that they could for small categories of goods across many discontinue publication of these statistics at the metropolitan areas. Therefore, any measure that national level, posing some risk for Option C. requires statistics on the prices of a small set Statistics on Prices of Broad Product of goods in metropolitan areas—which includes Categories Are Most Reliable. Statistics that options A, B, and C—may be at higher risk to future require obtaining representative price data for only BLS publication changes. However, the remaining broad categories of spending—such as all goods categories—food, apparel, transportation, and and services bought by a typical consumer— rent—include a much broader set of goods than the are likely to be the most reliable. Unlike the discontinued category. While we cannot predict the other options presented here, and the current future publication decisions of the BLS, it will be CNI, option D does not rely on statistics on any easier for the statistical agencies to obtain reliable specific spending categories. Instead, it uses data on prices of goods in these categories than only metropolitan area statistics on all goods and it will be for household fuels and other utilities, services purchased by a broad set of consumers. meaning it is more likely they will continue to publish Therefore, the BLS will likely continue publishing these statistics. the statistics required to calculate these statistics, making this option likely more robust to future publication decisions. RECOMMENDATION Rank of Options. To summarize, in Figure 3 on just on statistics for specific categories but also the next page, we provide relative assessments of on a national estimate for a category for which the each option using our criteria, dividing them into BLS has signaled difficulty obtaining reliable data; “green” for the best measures on each criterion, however, we believe that this risk is not significantly “yellow” for measures where some deficiency is higher than for options A and B. identified, and “red” for measures we judge to be Recommend Using National Estimate for significantly deficient. Options A, B, and C, which Discontinued Category. Overall, we think that largely maintain the current CNI calculation, are all option C, which uses a national estimate instead similar to the current CNI historically. By maintaining of metropolitan area estimates for the household an estimate for household fuels and other utilities, fuels and other utilities category, is best able option C has the most fidelity to the current CNI to replicate the current CNI. Figure 3 provides concept. Finally, option D, which does not use our overall ranking of each measure across the data on specific categories of goods, is likely the three criteria. Option C ranks best in both historic most immune from future publication changes. similarity and fidelity to the current concept. While Option C faces the most risk because it relies not it faces some risk from future publication decisions, www.lao.ca.gov 5 AN LAO REPORT these risks are similar to the risk Figure 3 for options A and B, which still require metropolitan area statistics Overall Rankings on prices for specific spending categories. Option D, which Measure Overall Historic Fidelity to Limited Future reflects, in part, changes in prices Rank Similarity Original Concept Changes of luxury goods and services, does not sufficiently capture the current Option A: 2 Remove CNI concept and is not historically Discontinued similar enough. Category Change Required For Option B: 3 2026-27 Budget. Under current Reassign statute, the CNI required for Expenditure Weights the 2026-27 budget will not be calculable based on published Option C: 1 statistics. Given that expenditure Use National Estimate estimates for the budget require use of the CNI, we recommend Option D: 4 that the Legislature enact a Use Alternative Index replacement option in statute as part of the 2026 budget process. LAO PUBLICATIONS This report was prepared by Alex Bentz, and reviewed by Brian Uhler and Carolyn Chu. The Legislative Analyst’s Office (LAO) is a nonpartisan office that provides fiscal and policy information and advice to the Legislature. To request publications call (916) 445-4656. This report and others, as well as an e-mail subscription service, are available on the LAO’s website at www.lao.ca.gov. The LAO is located at 925 L Street, Suite 1000, Sacramento, California 95814. 6 LEGISLATIVE ANALYST’S OFFICE