CMD researchers often receive questions about the Survey of Consumer Expectations (SCE) and its five survey modules (SCE Credit Access, SCE Household Spending, SCE Housing, SCE Labor Market, and SCE Public Policy).
Key characteristics:
Key characteristics:
- While the main SCE survey is fielded every month, the five special survey modules are fielded at a lower frequency. The SCE Credit Access, SCE Household Spending, SCE Public Policy, and SCE Labor Market surveys are fielded three times a year, while the SCE Housing Survey is fielded once a year.
- The sample of respondents for the SCE special modules is the same as that used for the SCE core survey and is designed to provide results that are nationally representative.
- First-time respondents of the SCE core survey in a given month do not receive the supplemental surveys that are fielded for that particular month. The SCE Housing Survey is run as a separate survey using the same sample population as that for the SCE core survey but is fielded separately from the SCE core survey.
SURVEY OF CONSUMER EXPECTATIONS (CORE) | SCE CREDIT ACCESS | SCE HOUSEHOLD SPENDING | SCE HOUSING | SCE LABOR MARKET | SCE PUBLIC POLICY |
When did the Survey of Consumer Expectations (SCE) first launch?
The survey officially launched in June 2013. Please refer to the corresponding Liberty Street Economics post for more information.
What is the sample size of the SCE?
The SCE is a nationally representative, internet-based survey of a rotating panel of approximately 1,300 household heads. Respondents participate in the panel for up to twelve months, with a roughly equal number rotating in and out of the panel each month. More information on the sample design can be found in the overview of the SCE methodology (see page 57).
When is the data collected each month?
The goal of the survey is to capture consumers’ expectations throughout a given month. The survey is structured so that a roughly equal amount of data are collected every week of the month. This is explained in more detail in the overview of the SCE methodology (see page 57).
Where can I find the survey questionnaire, and explanations of measures shown in the interactive charts?
Both can be found under the “downloads” pulldown on the main SCE interactive charts page as well in the Data Bank.
How does the survey define short-term inflation expectations versus medium- and longer-term inflation expectations?
With respect to inflation expectations, the short term is defined as one year ahead, the medium term as three years ahead, and the longer term as five years ahead. The three- and five-year questions ask respondents about the one-year-ahead change two and four years from now, respectively.
In the interactive charts and survey summary reports, why do you report the “median” for some variables and the “mean” for others?
The median is a robust measure of central tendency that is less sensitive to the presence of outliers than the mean (the arithmetic average). As it is our practice to not remove or top-code any outliers, for most expectations we report and focus on the median, with the exception of probabilistic expectations, for which we report the mean. For example, from the SCE core survey, we report the mean probability of losing or leaving a job, and the mean probability of finding a job. Generally, if the responses have bounds (that is, for job finding or job loss expectations they can respond with a number between 0-100), the survey reports the mean. Otherwise, the survey reports the median.
As most respondents typically provide integer answers, why are your reported medians not integer numbers?
As is common practice in case of integer variables, we compute an interpolated median. For more details, see page 65 in the overview of the SCE methodology.
What is the difference between the median expected inflation rate and the median point prediction?
The expected inflation rate is calculated using an individual’s density forecasts, obtained by asking each survey participant to assign probabilities to intervals of future inflation outcomes. More specifically, respondents are asked for the percent chance that, over the next twelve months, the rate of inflation (or deflation) will be 12 percent or higher; between 8 percent and 12 percent; between 4 percent and 8 percent; between 2 percent and 4 percent; or between 0 percent and 2 percent. Each individual’s responses are then used to fit a parametric density function (see page 56 of the overview of the SCE methodology), the mean of which is computed. The density mean for each respondent is included as a derived variable in the microdata we release. The median expected inflation rate is the median of those density means. The respondent’s point prediction instead is obtained by directly asking what the respondent thinks the inflation rate will be over a specified period (the next twelve months or, for example, during the period of May 2027 and May 2028). The median point forecast is the (interpolated) median of the respondents’ point forecasts. As earlier discussed by Engelberg, Manski, and Williams (JBES, 2009), respondents often report point forecasts that differ from their density means.
Why does the survey directly ask consumers for the rate of “inflation” versus expected changes in “prices,” as in other surveys?
The survey solicits responses on expectations about “inflation” because research suggests that asking about “prices” versus “inflation” can induce mixed interpretations, with some respondents thinking about specific prices they pay and others thinking about the overall rate of inflation. Asking directly about inflation yields more reliable, interpersonally comparable responses and is also consistent with the concept of forward inflation expectations, which are of interest to central banks. For further details, see the overview of the SCE methodology (page 55).
Do you have any detail on the variation in expectations across different regions?
The microdata permit analysis of differences in results by state, Census region, and Commuting Zones (CZ). Our interactive charts report findings for each of the four Census regions.
In reporting monthly SCE results, how do you define a record “low” or record “high” according to the survey?
For the SCE core, since the survey was first fielded in June 2013, a “lowest” or “highest” level on record is defined by the start of the survey in 2013. Other surveys on consumers’ expectations might have been fielded over a longer period of time, thus when the SCE cites records, they are according to the June 2013 start date. However, for other SCE special survey modules, there are different start dates for questions/series in each module.
Where can I find the microdata and when are the microdata released?
The microdata are publicly available for download here. The lag time for the microdata differs depending on the survey module. For the SCE core survey and SCE Credit Access module, microdata are posted with a nine-month lag. For the SCE Household Spending, SCE Public Policy, SCE Housing, and SCE Labor Market modules, microdata are posted with an eighteen-month lag.
What are the SCE “special modules” and how often are those surveys fielded?
Each month the SCE contains a supplementary “ad hoc” module on a special topic. There are four such modules repeated every four months. The four repeating supplements are on credit access, household spending, the labor market, and public policy. The SCE Housing Survey is fielded on an annual basis in February. The SCE Job Search module is also fielded on an annual basis, and the microdata are available in the Data Bank.
When did the SCE Credit Access Survey first launch?
Results from the SCE Credit Access Survey were first released in November 2014. The survey, however, has been fielded since October 2013. Please refer to the corresponding Liberty Street Economics post for more information.
How often is the survey fielded?
The SCE Credit Access Survey is fielded every four months, with data collected during February, June, and October of each year.
Where can I find the survey questionnaire and explanations of measures shown in the interactive charts?
Both can be found under the “downloads” pulldown on the SCE Credit Access webpage and in the Data Bank.
How is the “application rate” calculated?
All of the application rates are calculated as the weighted share of respondents who reported having applied for a form of credit (such as a credit card, auto loan, credit card limit increase, etc.), using the SCE monthly weights. For the broader measure—“applied to one or more types of credit”—it is the weighted share of respondents who reported having applied to at least one of the specified categories of credit (each of which has a separately reported application/request rate). For definitions of other measures, see the interactive chart guide.
When did the SCE Household Spending Survey first launch?
Results from the SCE Household Spending Survey were first released in February 2019. The survey, however, has been fielded since December 2014. Please refer to the corresponding Liberty Street Economics post for more information.
How often is the survey fielded?
The SCE Household Spending survey is fielded every four months, with data collected during April, August, and December of each year.
Where can I find the survey questionnaire and explanations of measures shown in the interactive charts?
Both can be found under the “downloads” pulldown on the SCE Household Spending webpage and in the Data Bank.
How do you define median year-ahead growth in overall household spending?
Median year-ahead growth is the median expected increase in household spending over the next twelve months. For definitions of other measures, see the interactive chart guide.
When did the SCE Housing Survey first launch?
Results from the SCE Housing Survey were first released in September 2014. The survey has been fielded since February 2014. Please refer to the corresponding Liberty Street Economics post for more information.
Why is the SCE Housing Survey only fielded on an annual basis?
Some housing market expectations are elicited in the monthly SCE core survey. However, as changes in the housing market tend to be more stable over time, this survey is fielded only in February of each year, providing an opportunity to take a more in-depth stock of homeowner and renter expectations and their correlates.
Where can I find the survey questionnaire and explanations of measures shown in the interactive charts? Both can be found under the “downloads” pulldown on the SCE Housing Survey webpage and in the Data Bank.
Does the SCE Housing Survey address the expectations and concerns of renters as well as homeowners?
Yes, questions in the survey are fielded to both renters and owners, with questions specific to each group.
Do you provide any regional or state-specific data?
The survey collects information on demographic characteristics including age, gender, and residential location. The microdata permit analysis of differences in results by state, Census region, and Commuting Zones (CZ). Our interactive charts report findings only for each of the four Census regions for which the survey is nationally representative (Midwest, Northeast, South, and West). All statistics are weighted to be representative of the U.S population of household heads in terms of income, education, age, and Census region. For definitions of other measures, see the interactive chart guide.
When did the SCE Labor Market Survey first launch?
Results from the SCE Labor Market Survey were first released in August 2017. The survey, however, has been fielded since March 2014. Please refer to the corresponding Liberty Street Economics post for more information.
How often is the survey fielded?
The SCE Labor Market survey is fielded every four months, with data collected during March, July, and November of each year.
Where can I find the survey questionnaire, and explanations of measures shown in the interactive charts?
Both can be found under the “downloads” pulldown on the SCE Labor Market webpage and in the Data Bank.
How do you interpret the expectations of working beyond a certain age?
The survey asks for the percent chance of working beyond a certain age, not whether respondents will work beyond a certain age or not. For example, we may summarize results in a press release as: “The average expected likelihood of working beyond age 62 declined to 49.3 percent.” The correct interpretation of the data is “the average probability respondents assign to working beyond age 62 is 49.3 percent.” For definitions of other measures, see the interactive chart guide.
When did the SCE Public Policy Survey first launch?
Results from the SCE Public Policy Survey were first released in October 2019. The survey, however, has been fielded since November 2015. Please refer to the corresponding Liberty Street Economics post for more information.
What are the goals of this survey?
The SCE Public Policy Survey provides a comprehensive picture of consumers’ expectations regarding a set of public policy changes related to taxes and entitlement programs. Although many surveys, including the SCE, measure households’ expectations about future economic outcomes for themselves and for the economy overall, surprisingly little is known about expectations regarding future policy changes at the local and federal levels, especially in relation to taxes and entitlement programs. These data can help shed light on how government policy actions influence the dynamics and near-term trajectory of consumer spending, work, and savings behavior in real time.
How often is the survey fielded?
The SCE Public Policy Survey is fielded every four months, with data collected during April, August, and December of each year (with the exception of the first survey, which was fielded in November).
Where can I find the survey questionnaire, and explanations of measures shown in the interactive charts?
Both can be found under the “downloads” pulldown on the SCE Public Policy webpage and in the Data Bank.
Updated January 2023
The survey officially launched in June 2013. Please refer to the corresponding Liberty Street Economics post for more information.
What is the sample size of the SCE?
The SCE is a nationally representative, internet-based survey of a rotating panel of approximately 1,300 household heads. Respondents participate in the panel for up to twelve months, with a roughly equal number rotating in and out of the panel each month. More information on the sample design can be found in the overview of the SCE methodology (see page 57).
When is the data collected each month?
The goal of the survey is to capture consumers’ expectations throughout a given month. The survey is structured so that a roughly equal amount of data are collected every week of the month. This is explained in more detail in the overview of the SCE methodology (see page 57).
Where can I find the survey questionnaire, and explanations of measures shown in the interactive charts?
Both can be found under the “downloads” pulldown on the main SCE interactive charts page as well in the Data Bank.
How does the survey define short-term inflation expectations versus medium- and longer-term inflation expectations?
With respect to inflation expectations, the short term is defined as one year ahead, the medium term as three years ahead, and the longer term as five years ahead. The three- and five-year questions ask respondents about the one-year-ahead change two and four years from now, respectively.
In the interactive charts and survey summary reports, why do you report the “median” for some variables and the “mean” for others?
The median is a robust measure of central tendency that is less sensitive to the presence of outliers than the mean (the arithmetic average). As it is our practice to not remove or top-code any outliers, for most expectations we report and focus on the median, with the exception of probabilistic expectations, for which we report the mean. For example, from the SCE core survey, we report the mean probability of losing or leaving a job, and the mean probability of finding a job. Generally, if the responses have bounds (that is, for job finding or job loss expectations they can respond with a number between 0-100), the survey reports the mean. Otherwise, the survey reports the median.
As most respondents typically provide integer answers, why are your reported medians not integer numbers?
As is common practice in case of integer variables, we compute an interpolated median. For more details, see page 65 in the overview of the SCE methodology.
What is the difference between the median expected inflation rate and the median point prediction?
The expected inflation rate is calculated using an individual’s density forecasts, obtained by asking each survey participant to assign probabilities to intervals of future inflation outcomes. More specifically, respondents are asked for the percent chance that, over the next twelve months, the rate of inflation (or deflation) will be 12 percent or higher; between 8 percent and 12 percent; between 4 percent and 8 percent; between 2 percent and 4 percent; or between 0 percent and 2 percent. Each individual’s responses are then used to fit a parametric density function (see page 56 of the overview of the SCE methodology), the mean of which is computed. The density mean for each respondent is included as a derived variable in the microdata we release. The median expected inflation rate is the median of those density means. The respondent’s point prediction instead is obtained by directly asking what the respondent thinks the inflation rate will be over a specified period (the next twelve months or, for example, during the period of May 2027 and May 2028). The median point forecast is the (interpolated) median of the respondents’ point forecasts. As earlier discussed by Engelberg, Manski, and Williams (JBES, 2009), respondents often report point forecasts that differ from their density means.
Why does the survey directly ask consumers for the rate of “inflation” versus expected changes in “prices,” as in other surveys?
The survey solicits responses on expectations about “inflation” because research suggests that asking about “prices” versus “inflation” can induce mixed interpretations, with some respondents thinking about specific prices they pay and others thinking about the overall rate of inflation. Asking directly about inflation yields more reliable, interpersonally comparable responses and is also consistent with the concept of forward inflation expectations, which are of interest to central banks. For further details, see the overview of the SCE methodology (page 55).
Do you have any detail on the variation in expectations across different regions?
The microdata permit analysis of differences in results by state, Census region, and Commuting Zones (CZ). Our interactive charts report findings for each of the four Census regions.
In reporting monthly SCE results, how do you define a record “low” or record “high” according to the survey?
For the SCE core, since the survey was first fielded in June 2013, a “lowest” or “highest” level on record is defined by the start of the survey in 2013. Other surveys on consumers’ expectations might have been fielded over a longer period of time, thus when the SCE cites records, they are according to the June 2013 start date. However, for other SCE special survey modules, there are different start dates for questions/series in each module.
Where can I find the microdata and when are the microdata released?
The microdata are publicly available for download here. The lag time for the microdata differs depending on the survey module. For the SCE core survey and SCE Credit Access module, microdata are posted with a nine-month lag. For the SCE Household Spending, SCE Public Policy, SCE Housing, and SCE Labor Market modules, microdata are posted with an eighteen-month lag.
What are the SCE “special modules” and how often are those surveys fielded?
Each month the SCE contains a supplementary “ad hoc” module on a special topic. There are four such modules repeated every four months. The four repeating supplements are on credit access, household spending, the labor market, and public policy. The SCE Housing Survey is fielded on an annual basis in February. The SCE Job Search module is also fielded on an annual basis, and the microdata are available in the Data Bank.
SURVEY OF CONSUMER EXPECTATIONS (CORE) | SCE CREDIT ACCESS | SCE HOUSEHOLD SPENDING | SCE HOUSING | SCE LABOR MARKET | SCE PUBLIC POLICY |
When did the SCE Credit Access Survey first launch?
Results from the SCE Credit Access Survey were first released in November 2014. The survey, however, has been fielded since October 2013. Please refer to the corresponding Liberty Street Economics post for more information.
How often is the survey fielded?
The SCE Credit Access Survey is fielded every four months, with data collected during February, June, and October of each year.
Where can I find the survey questionnaire and explanations of measures shown in the interactive charts?
Both can be found under the “downloads” pulldown on the SCE Credit Access webpage and in the Data Bank.
How is the “application rate” calculated?
All of the application rates are calculated as the weighted share of respondents who reported having applied for a form of credit (such as a credit card, auto loan, credit card limit increase, etc.), using the SCE monthly weights. For the broader measure—“applied to one or more types of credit”—it is the weighted share of respondents who reported having applied to at least one of the specified categories of credit (each of which has a separately reported application/request rate). For definitions of other measures, see the interactive chart guide.
SURVEY OF CONSUMER EXPECTATIONS (CORE) | SCE CREDIT ACCESS | SCE HOUSEHOLD SPENDING | SCE HOUSING | SCE LABOR MARKET | SCE PUBLIC POLICY |
When did the SCE Household Spending Survey first launch?
Results from the SCE Household Spending Survey were first released in February 2019. The survey, however, has been fielded since December 2014. Please refer to the corresponding Liberty Street Economics post for more information.
How often is the survey fielded?
The SCE Household Spending survey is fielded every four months, with data collected during April, August, and December of each year.
Where can I find the survey questionnaire and explanations of measures shown in the interactive charts?
Both can be found under the “downloads” pulldown on the SCE Household Spending webpage and in the Data Bank.
How do you define median year-ahead growth in overall household spending?
Median year-ahead growth is the median expected increase in household spending over the next twelve months. For definitions of other measures, see the interactive chart guide.
SURVEY OF CONSUMER EXPECTATIONS (CORE) | SCE CREDIT ACCESS | SCE HOUSEHOLD SPENDING | SCE HOUSING | SCE LABOR MARKET | SCE PUBLIC POLICY |
When did the SCE Housing Survey first launch?
Results from the SCE Housing Survey were first released in September 2014. The survey has been fielded since February 2014. Please refer to the corresponding Liberty Street Economics post for more information.
Why is the SCE Housing Survey only fielded on an annual basis?
Some housing market expectations are elicited in the monthly SCE core survey. However, as changes in the housing market tend to be more stable over time, this survey is fielded only in February of each year, providing an opportunity to take a more in-depth stock of homeowner and renter expectations and their correlates.
Where can I find the survey questionnaire and explanations of measures shown in the interactive charts? Both can be found under the “downloads” pulldown on the SCE Housing Survey webpage and in the Data Bank.
Does the SCE Housing Survey address the expectations and concerns of renters as well as homeowners?
Yes, questions in the survey are fielded to both renters and owners, with questions specific to each group.
Do you provide any regional or state-specific data?
The survey collects information on demographic characteristics including age, gender, and residential location. The microdata permit analysis of differences in results by state, Census region, and Commuting Zones (CZ). Our interactive charts report findings only for each of the four Census regions for which the survey is nationally representative (Midwest, Northeast, South, and West). All statistics are weighted to be representative of the U.S population of household heads in terms of income, education, age, and Census region. For definitions of other measures, see the interactive chart guide.
SURVEY OF CONSUMER EXPECTATIONS (CORE) | SCE CREDIT ACCESS | SCE HOUSEHOLD SPENDING | SCE HOUSING | SCE LABOR MARKET | SCE PUBLIC POLICY |
When did the SCE Labor Market Survey first launch?
Results from the SCE Labor Market Survey were first released in August 2017. The survey, however, has been fielded since March 2014. Please refer to the corresponding Liberty Street Economics post for more information.
How often is the survey fielded?
The SCE Labor Market survey is fielded every four months, with data collected during March, July, and November of each year.
Where can I find the survey questionnaire, and explanations of measures shown in the interactive charts?
Both can be found under the “downloads” pulldown on the SCE Labor Market webpage and in the Data Bank.
How do you interpret the expectations of working beyond a certain age?
The survey asks for the percent chance of working beyond a certain age, not whether respondents will work beyond a certain age or not. For example, we may summarize results in a press release as: “The average expected likelihood of working beyond age 62 declined to 49.3 percent.” The correct interpretation of the data is “the average probability respondents assign to working beyond age 62 is 49.3 percent.” For definitions of other measures, see the interactive chart guide.
SURVEY OF CONSUMER EXPECTATIONS (CORE) | SCE CREDIT ACCESS | SCE HOUSEHOLD SPENDING | SCE HOUSING | SCE LABOR MARKET | SCE PUBLIC POLICY |
When did the SCE Public Policy Survey first launch?
Results from the SCE Public Policy Survey were first released in October 2019. The survey, however, has been fielded since November 2015. Please refer to the corresponding Liberty Street Economics post for more information.
What are the goals of this survey?
The SCE Public Policy Survey provides a comprehensive picture of consumers’ expectations regarding a set of public policy changes related to taxes and entitlement programs. Although many surveys, including the SCE, measure households’ expectations about future economic outcomes for themselves and for the economy overall, surprisingly little is known about expectations regarding future policy changes at the local and federal levels, especially in relation to taxes and entitlement programs. These data can help shed light on how government policy actions influence the dynamics and near-term trajectory of consumer spending, work, and savings behavior in real time.
How often is the survey fielded?
The SCE Public Policy Survey is fielded every four months, with data collected during April, August, and December of each year (with the exception of the first survey, which was fielded in November).
Where can I find the survey questionnaire, and explanations of measures shown in the interactive charts?
Both can be found under the “downloads” pulldown on the SCE Public Policy webpage and in the Data Bank.
Updated January 2023
