2022 Consumer Price Index on a Local Scale and Public Perception of Inflation

Abstract

The Penn State Economics Association Research Team has worked to model a consumer price index for the average Penn State student during the 2022 Fall Semester. This research draws inspiration from Kansas State’s 2020 Consumer Price Index, and builds on last semester’s consumer price index constructed by the team. This updated price index seeks to highlight the ways in which student spending has changed in the wake of inflation and increased housing prices. The research team conducted a ‘perceptions of inflation’ survey as well as a CPI survey in order to both understand student views on inflation and to construct a market basket of goods based on student spending habits. In addition to survey administration, the research team gathered prices of numerous goods and services across the borough. The Fall 2022 consumer price index seeks to track changes in student spending habits in the wake of this unique economic landscape and stand as a basis of comparison for previous and subsequent semesters. The development of a student CPI is beneficial to both the student body and the greater State College community, which seeks to foster an environment that supports both local businesses and students.

Introduction

As the Federal Reserve continues to tighten monetary policy to bring down the PCE inflation rate back to its target of 2 percent, the US economy has already felt the effect of rising interest rates. The Penn State Economics Association Research Team set out to gather data about how localized prices in State College, Pennsylvania have changed for students.

The data we compiled is known as the Student Price Index (SPI) which is based on prices collected on a bundle of goods a typical Penn State University student purchases. The Consumer Price Index (CPI) is a measure of the average change over time in the prices paid by consumers for a market basket of consumer goods and services. A market basket of goods refers to a selection of goods and services that are consistently purchased and sold throughout an economic system. Economists, politicians, and financial analysts use market baskets as a way to track price changes over time and determine inflation levels.

We will use the SPI to compare differences in price fluctuations between our localized student basket and the national average. By collecting this data, we hope to set a precedent for future years to compute their data with to observe the change in prices for students in State College. A similar study was previously conducted by Kansas State University and we hope to replicate their procedure here in our own localized area. We set out to answer this question: How much were State College students affected by national inflation at the local level?

Methodology

To gather the necessary data about students’ consumption of goods, a survey of students’ spending on certain items and where they purchase these items had to be developed. The critical components of the survey had to gather the monetary value spent on each category: Dining/takeout, gas, tuition, alcohol, non-Greek housing, textbooks, utilities, and groceries. Additionally, identifying the specific amounts of money spent per category and the most common place where those items were purchased was important to produce viable results.

Population and Distribution

The population that the survey was aimed at was specifically Pennsylvania State students who were undergraduates and between the ages of 18-24, this being a rough estimate of the average age range of students studying for an undergraduate degree. To reach this population, the survey was distributed by 2 distribution devices that incorporated three methods. The most common device used for the survey was through an anonymous link that was sent out to extracurricular group chats or organizations. This method accounted for 59 (78%) respondents. The next distribution device included two different methods of distribution to reach the population. This device was a QR code and this image was either sent in group chats by members of the research team or it was distributed physically, either by a handout or being posted around campus. Through the QR, 17 (22%) responses were collected. Overall, the distribution methods allowed the survey to reach multiple student organizations and classes.

Device

Another importance when it came to obtaining as many responses as possible was to have the survey clean and presentable. The best way to achieve this was to use Qualtrics as the survey delivery device. Qualtrics allows for complete user control of survey questions with format, question-type, response restrictions, logic-based survey flow, etc. It allowed for the research team to format the questions in the most understandable way for the respondent and also allowed for easy exportation of the data and other information to run the best analysis on it.

Questions

The survey first asked respondents for a Penn State email as a way to verify the respondent’s enrollment in Penn State. The next few questions characterized spending activity by asking if the individual lived on campus, if they pay for a Bus pass, if they have a meal plan with the university, and if they purchase Penn State Sports tickets. These questions characterize possible contributing factors for certain categories of the survey. For instance, having a meal plan might affect the amount of money spent on food or having a bus pass eliminates the need to purchase gas. After this block of questioning, the survey turns towards monetary questions about each category within the SPI’s basket, questions that ask the respondent to rank the categories, and where they most often purchase the given categories.

Calculations

Using the data derived from the SPI survey and raw data collected through the team's work, the quantitative team conducted four calculations that would analyze the results of the survey data and the price data that was compiled. These calculations retrieved and compared the data to both previous local data and national data. The first calculation directly found the value of the SPI using the data collected from local businesses about the prices of products. This calculation was the focus of the research and the basis for further calculations. The second calculation that was conducted used the survey data compiled through the SPI survey. The goal was to estimate the proportions of spending that students conducted across the various sectors of the SPI. This calculation allowed the quantitative team to compare the results from the SPI survey and the price data collected from businesses to data collected from the Kansas State University Economics Club and previous data collected from the Penn State Economics Association (PSUEA). The third calculation compared the inflation rate in each sector from the data collected in the Spring semester of 2022.

The final calculation compared all of the data collected to previously collected data. Price data between products from Spring 2022 to Fall 2022, SPI comparisons between Spring 2022 and Fall 2022 at PSU, comparisons between college campuses, and a comparison to national CPI data. This calculation allowed for quality analysis upon the results to be conducted and gave an insight into the differences in prices and peoples’ spending depending on location, time, and position in life. Detailed methodology on each calculation is given below.

Calculation 1

The first calculation was arguably the most important for this project. This was a simple consumer price index calculation in which the total inflation rate across State College was measured using data on consumption expenditures between Fall and Spring 2022. The first step was to create a shared basket of goods for which we had data from both semesters. In order to do this, we removed all goods which weren’t recorded in both datasets so we could ensure we had accurate comparisons of changes in price levels. We then calculated the total expenditures for the average student on the goods remaining in the basket across both semesters to find the CPI for each semester. The final step was to calculate the inflation rate using the standard CPI formula.

Calculation 2

Calculating spending levels across each sector was quite simple. Like the other calculations, we started by cleaning the data. This involved removing entries where consumption estimates were provided as ranges or where not all entries were properly recorded. After this was done, we created two estimates. The first was a measure of total recorded expenditures on each sector. To calculate this, we summed the total spending per sector and divided it by the total spending recorded in the survey. The second estimate was created by calculating the percentage of income which each respondent devoted to any particular sector and finding the average across all respondents. Both of these metrics are useful in their own ways: The first gives us an idea of where money is flowing across all of State College’s industries, while the second gives us an idea of individual students’ spending habits.

Calculation 3

The purpose of the third calculation, as stated previously, was to compare changes in prices across the various sectors between Fall and Spring 2022. In order to do this, we had to make some mutations to the spring survey data in order to account for differing questions between the two surveys. This included multiplying all weekly expenses collected in the spring survey by a factor of 4.35 in order to estimate monthly expenses. We also eliminated clothing from the basket of goods being assessed, as it was not considered under the fall 2022 survey. Finally, we removed expenses on tuition costs from the fall 2022 dataset, as those were not recorded in the spring survey.

After cleaning the data, two separate datasets were created: the first contained all student responses, and the second contained only the responses of students who reported expenditures in every service. This was done so that we could analyze the raw change in prices for each sector without the data being skewed by individuals who simply didn’t use those services.

For each dataset, summary statistics were calculated, and we measured the change in mean and median expenditures for each sector using the standard CPI formula. These changes were then graphed in a two-column bar chart.

Calculation 4

The final calculation was essentially a summary of all previous calculations. We gathered changes in CPI across Penn State from Calculations 1 and 3, as well as changes in spending patterns from Calculation 2 and compared them to Spring period data, as well as data from a number of other college campuses both within and out of Penn State, as well as the national inflation rate.

Results

The results can be broken down into two categories: spending patterns, and price change comparisons. Despite revolving around the same focus of State College, these shed light on different aspects of the current economy.

Spending Patterns

The SPI survey was used to produce a breakdown of monthly spending by local Penn State students. The team chose to display a breakdown of food spending patterns separately from the overall breakdown. It was found that over half of the average student’s monthly budget, 57.1%, went to rent for housing. This monetarily translates into $1,078.48. The second-biggest expense was through the university-provided meal plan at $240.71 or 12.7%. Similarly, following the meal plan was off-campus food at $182.06 or 9.6%, After the off-campus food comes the groceries with an amount of $151.55 spent each month which is 8.0 % of the average monthly spendings of Penn State students. Furthermore, Alcohol represented 5.8 % which is about $109.47 monthly. Finally, purchasing Textbooks ranked with the lowest percentage at a percentage of 3.1% in other words only $58.02 is spent on Textbooks.

The survey separated food spending into seven categories; Chinese, Pizza, Fast Food, Sandwiches, American, Japanese and Mexican. It was found that 33.7% of the food breakdown was spent on Fast Food, monetarily that is around $64.43. Mexican food ranked second highest with 16.5% which is $31.42, then American food with a percentage of 14.9% or $28.42. Similarly, with a small percentage difference Sandwiches has 13.4% which is monetarily $25.55, After the Sandwiches comes the Pizza with 11.3% or $21.55. Furthermore, the Chinese food came in before the last, with 6.6% which is around $12.52, and finally the Japanese food ranked in the last place, with a percentage of 3.7% which equates to $7.10.

Figure 2: Student Food Spending Patterns

Price Change Comparisons

This study directly follows a similar study conducted the semester prior in Spring of 2022. A local Consumer Price Index was calculated then and is now set as a benchmark to calculate the direction and magnitude of the change in local prices for which no official reports exist. The comparison between the two calculations shows a -5.7% change in local prices between March of 2022 and October of 2022. This is due to the -$75 dollar change in the chosen basket of goods as it went from $1395 in the Spring of 2022 to $1320 in the Fall of the same calendar year.

The impact of each sector’s change in price varied based on whether mean or median expenditures were used. Using mean prices, the majority of disinflation seems to have been led by a decrease in gas and housing prices, which fell by 30.55% and 23.64% respectively. Meanwhile, using median prices, the majority of disinflation appears to have occurred from changes in alcohol, textbook, and gas expenditures, which were recorded to have fallen by 61.69%, 32.53%, and 31.03% respectively.

There were some significant disparities in price changes of certain sectors depending on whether mean or median were used. For instance, utilities expenses appeared to increase by 32.85% using mean expenditures, whereas they fell by 16.67% using median expenditures. Similarly, under mean expenditures, alcohol expenses only fell by 8.29% as opposed to the aforementioned 61.69% using median expenditures. This indicates the presence of a skew in the data, even accounting for outliers, among these sectors.

Figure 5: Comparison of Sector Expenditures

Discussion

Interpreting Results

After comparing the price change between the spring and the fall of 2022, it has been noticed that the prices of goods and services have fallen by 5.7% from the spring to the fall semester of the year 2022, as the chosen basket of goods went from $1395 to $1320 in the fall of 2022. This implies that the purchasing power of students has increased over that time period. When discussing the spending patterns, the data showed that the highest amount of students' money is spent on rent, as rent represents 57.1% of the student’s average monthly spending. Alcohol comes in last place with a percentage of 3.1%. When comparing these two percentages with the Kansas State University student SPI, we found that the rankings differ significantly. Rent only represents 6.1% while Alcohol represents 16.2% of the total average spending of Kansas State University.

Public Perception of Prices

Country wide, inflation has been a topic of concern for many consumers. The two periods which this study looked at were the spring and fall of 2022. This is a year in which the inflation rate has been well above its target rate globally, nationally, and often locally. The research team conducted an initial survey to better understand perceptions of price changes. Penn State students were asked whether they believed prices had gone up or down during the year of 2022. Every respondent either believed prices had gone up or gave a neutral response.

Our findings run contrary to perceptions. A decrease in prices of 5.7% throughout 2022 is in many ways unexpected. It also is unexplained, and further study would be needed to understand why prices decreased in this manner. However, the public's perceptions of rising prices often do not reflect the real situation at local levels. Localized inflation can fluctuate a great deal around the country wide average rate.

Strengths and Limitations

The most valuable asset to this study was the experience of its team members. We had two main advantages. The first being that all of us were economics majors, meaning we had relevant experience in reading and designing studies around this topic. Secondly, all researchers are students. This meant we had a general idea of the nature of student spending on campus, as well as what sectors we should ask for expenditures on.The wide array of skill sets on our team allowed us to conduct multiple complex analyses using tools such as R, Excel, and Matlab, as well as properly convey our results and findings through our presentation and through this report.

We are incredibly proud of the work we’ve done so far, but we also acknowledge that there were certain flaws and limitations with this project that must be improved upon in the future. Arguably the largest limitation of this study was the lack of a sufficient sample size. While the fall 2022 survey received more responses than the one conducted in the spring, the sample still only amounted to 90 responses, 14 of which had to be eliminated during the cleaning stage due to unclear responses. Finding methods of distributing the survey across a wider sample and increasing the clarity and uniformity of responses should be one of our biggest priorities with this project moving forward, especially if we wish to annually update the price index. This is a logistical issue because the nature of the project seems to require an extensive survey unlike the team’s newest risk aversion study which garnered 300+ responses in about half the time frame.

Additional limitations include a higher degree of organization and planning with respect to the survey design. Our analysis and design processes were heavily disconnected, with the study design occurring before we had sufficiently fleshed out the calculations and analyses we’d be performing. As such, certain questions from the survey failed to give us the data we needed outright and had to be heavily modified during the cleaning stage to make the analysis easier. This not only increased workload, but also raised the possibility of error in our findings due to imprecise transformations being conducted on the dataset. One possible solution to this is the standardization of the study going forward - the questions should ideally remain the same so they are as similar as possible to one-another when updating the SPI in the future. Any changes done to the survey should only be implemented after a fleshed out plan of analysis has been drafted.

Due to the survey’s use of free-response answers, another issue with data collection was the use of non-standard responses, such as ranges for expenditures in a given area. These kinds of responses were incredibly difficult to standardize and incorporate into the dataset, and ultimately had to be eliminated. By adopting a more robust and coherent response system, we could avoid this issue in the future and retain a larger percentage of our responses to include in the samples of our analyses. This would also allow for a more streamlined approach to data processing.

The final limitation worth mentioning was the difficulty coordinating the various calculations. Each calculation was assigned to individual team members, and there was little collaboration between members working on different calculations. This meant that there was no standardized methodology developed for cleaning and analyzing the data across the various analyses that were conducted, potentially leading to errors in calculations that set out to measure similar metrics (the prime examples being Calculation 1 and Calculation 3).

All of these issues give us a list of things we can improve on as we expand on this project moving forward, and we’re confident that they can be resolved in future surveys.

Further Study

This calculation of SPI can be conducted in future periods and improved upon. The PSUEA Research team will continue to refine the process by updating the survey and increasing distribution, calculating more efficiently, and setting up dedicated subgroups. As our research in this subject grows, so will understanding of localized inflation for Penn State students.

Future similar studies by other universities which focus on constructing a CPI basket for their respective student populations would contribute significantly to expanding this literature and refining the process of local price index research. It is crucially important to understand inflation and its effect on a local scale. Many individuals, especially college students, are accustomed to doing the majority of their spending in one area. Living, working, and spending in State College, PA means that the price fluctuations in that area matter most to these individuals.

The Economics of Student Discounts:

An Introduction to 3rd Degree Price Discrimination

When regular consumers are confronted with a particular good being sold by a particular firm they generally imagine that the price they pay is the price that every other consumer pays. If this is not the case then the firm is using nonuniform pricing. There are many reasons this can occur. For example, a consumers living in Alaska or Hawaii are accustomed to paying more than mainland Americans for some shipped goods due to increased marginal cost for the firm. However, when price variations are due to a firm with market power attempting to maximize profit, this is referred to as price discrimination.

Every consumer who buys a firm's product gets some consumer utility greater than or equal to zero, which can be calculated by subtracting the price they paid from the maximum price they would be willing to pay. Price discrimination can be thought of as a firm attempting to capture some of this surplus for themselves. In order to maximize profit, a firm will practice price discrimination if it is able to do so. A firm is able to price discriminate if three conditions are met: some market power, ability to identify willingness to pay, and ability to prevent or limit resale. If a firm has a great deal of market power, knows every consumer's maximum price, and can prevent all resale; the firm is able to practice 1st degree price discrimination or perfect price discrimination, where every consumer pays their maximum price and all consumer surplus is converted to producer surplus.

More commonly a firm can separate consumers into two or more groups, where it is known that each group has a different willingness to pay and therefore a different demand curve. If the firm also has some market power and ability to limit resale, a different profit maximizing price can be set for each group. This is known as 3rd degree price discrimination.

*3rd Degree Price Discrimination with Two Consumer Groups*

Student discounts are a very common form of 3rd degree price discrimination. In general, college students have much lower income than other adults. This, along with other factors, means they often have a different demand curve. While firms in the real world cannot find the exact nature of that difference, the firms can know enough to set prices lower for students and higher for other consumers (if they have the market power to do so). Students will notice and take advantage of this lower price to buy goods that they might not have otherwise bought if the price was set at the profit maximizing point for all consumer demand curves.

The implication, which many consumers don’t realize, is that if the firm practiced a uniform pricing structure the optimal price would always be lower than the non-student price. Anyone who is not a student and paying full price is having some of the consumer surplus they would have gained under uniform pricing taken by the firm. Firms do not have any incentive to give students a better deal for the sake of it; however, firms may also benefit from favorable publicity, or other less tangible benefits. When it comes to 3rd degree price discrimination, firms are driven by profit motives. Separating students from non-students can give them the ability to extract a higher revenue by optimizing price for each demand curve separately.

Sources

Carlton Dennis W and Jeffrey M Perloff. Modern Industrial Organization. 4th ed. Pearson/Addison Wesley 2005.

McLeod, Mark. ECON342: Industrial Organization. The Pennsylvania State University. Class Lecture.