Consumer Expenditure Survey - CX

Data - BLS

First choose demographics:

• Demographics: LB01 - By Quantile, LB02 - By Income

• Characteristics: for each demographic type (e.g. 01), list of groups. (All, 1st quantile etc.)

Then choose item:

• Category: EXPEND - Expenditures, or INCOME - Income and Taxes

• Subcategory:

• Item:

Last observation: 2024 (N = 96030)

First observation: 1984 (N = 10358)

Last data update: 23 sept. 2026, 23:33

Last compile: 23 sept. 2026, 23:40

Demographics

cx.demographics

cx.demographics %>%
  {if (is_html_output()) print_table(.) else .}

demographics_code	demographics_text	display_level	selectable	sort_sequence
LB01	Quintiles of income before taxes	0	T	100
LB02	Income before taxes	0	T	200
LB04	Age of reference person	0	T	300
LB05	Size of consumer unit	0	T	400
LB06	Composition of consumer unit	0	T	400
LB07	Number of earners	0	T	500
LB09	Race of reference person	0	T	700
LB10	Hispanic or Latino origin of reference person	0	T	800
LB11	Region of residence	0	T	900
LB12	Occupation of reference person	0	T	1000
LB13	Education of reference person	0	T	1100
LB14	Highest education level of any member	0	T	1200
LB15	Deciles of income before taxes	0	T	1300
LB16	Generation of reference person	0	T	1600
LB17	Housing tenure	0	T	1700
LB18	Type of area	0	T	1800
LB19	Type of area	0	T	1900
LB20	Population size of area of residence	0	T	2000
LB21	Selected age of reference person	0	T	2100

cx.characteristics

cx.characteristics %>%
  {if (is_html_output()) datatable(., filter = 'top', rownames = F) else .}

Item

cx.category

cx.category %>%
  {if (is_html_output()) print_table(.) else .}

category_code	category_text	display_level	selectable	sort_sequence
ADDENDA	Assets and liabilities, and other financial info	0	T	300
CUCHARS	Consumer Characteristics	0	T	400
EXPEND	Expenditures	0	T	100
INCOME	Income and Taxes	0	T	200

cx.subcategory

cx.subcategory %>%
  {if (is_html_output()) datatable(., filter = 'top', rownames = F) else .}

cx.item

cx.item %>%
  {if (is_html_output()) datatable(., filter = 'top', rownames = F) else .}

Income Example

By Quintile

cx.series |>
  filter(demographics_code == "LB01",
         item_code %in% c("INCAFTTX", "HOUSING", "INSPENSN", "TOTALEXP", "OWNMORTG")) |>
  inner_join(cx.data.1.AllData |>
              filter(year == 2016), by = "series_id") |>
  inner_join(cx.characteristics |>
              select(1, 2, 3), by = c("characteristics_code", "demographics_code")) |>
  select(characteristics_code, characteristics_text, item_code, value) |>
  spread(item_code, value) %>%
  {if (is_html_output()) print_table(.) else .}

characteristics_code	characteristics_text	HOUSING	INCAFTTX	INSPENSN	OWNMORTG	TOTALEXP
01	All Consumer Units	18886	64175	6831	2889	57311
02	Lowest 20 percent income quintile	10267	11832	645	545	25138
03	Second 20 percent income quintile	13552	29423	1766	977	36770
04	Third 20 percent income quintile	16315	47681	4227	1910	47664
05	Fourth 20 percent income quintile	20687	75065	8262	3707	64910
06	Highest 20 percent income quintile	33653	157215	19302	7325	112221

By Decile

cx.series |>
  filter(demographics_code == "LB15",
         item_code %in% c("INCAFTTX", "HOUSING", "INSPENSN", "TOTALEXP", "OWNMORTG")) |>
  inner_join(cx.data.1.AllData |>
              filter(year == 2016), by = "series_id") |>
  inner_join(cx.characteristics |>
              select(1, 2, 3), by = c("characteristics_code", "demographics_code")) |>
  select(characteristics_code, characteristics_text, item_code, value) |>
  spread(item_code, value) %>%
  {if (is_html_output()) print_table(.) else .}

characteristics_code	characteristics_text	HOUSING	INCAFTTX	INSPENSN	OWNMORTG	TOTALEXP
01	All consumer units	18886	64175	6831	2889	57311
02	Lowest 10 percent	9567	6774	644	512	23588
03	Second 10 percent	10961	16841	645	579	26675
04	Third 10 percent	12829	25423	1411	760	34221
05	Fourth 10 percent	14271	33404	2120	1193	39308
06	Fifth 10 percent	15511	42410	3290	1547	43975
07	Sixth 10 percent	17119	52949	5164	2272	51351
08	Seventh 10 percent	19285	66676	7058	3198	59395
09	Eighth 10 percent	22085	83424	9461	4215	70411
10	Ninth 10 percent	26719	108743	13278	5564	87432
11	Highest 10 percent	40547	205391	25290	9075	136873

By Income

cx.series |>
  filter(demographics_code == "LB02",
         item_code %in% c("INCAFTTX", "HOUSING", "INSPENSN", "TOTALEXP", "OWNMORTG")) |>
  inner_join(cx.data.1.AllData |>
              filter(year == 2016), by = "series_id") |>
  inner_join(cx.characteristics |>
              select(1, 2, 3), by = c("characteristics_code", "demographics_code")) |>
  select(characteristics_code, characteristics_text, item_code, value) |>
  spread(item_code, value) |>
  filter(!is.na(INCAFTTX)) %>%
  {if (is_html_output()) print_table(.) else .}

characteristics_code	characteristics_text	HOUSING	INCAFTTX	INSPENSN	OWNMORTG	TOTALEXP
01	All Consumer Units	18886	64175	6831	2889	57311
07	$30,000 to $39,999 before tax income	14533	34381	2222	1274	40144
08	$40,000 to $49,999 before tax income	15575	43047	3400	1554	44150
09	$50,000 to $69,999 before tax income	17331	54782	5307	2451	52088
18	Less than $15,000	9698	8732	614	551	23657
19	$15,000 to $29,999	12268	23012	1161	678	31913
20	$70,000 to $99,999	20564	74744	8270	3632	65086
21	$100,000 to $149,999	26003	103504	12543	5299	84154
22	$150,000 to $199,999	33319	140130	17609	7391	109516
23	$200,000 and more	46076	254890	31079	10333	158896

Example: All Items

cx.series |>
  filter(demographics_code == "LB15",
         characteristics_code == "01") |>
  inner_join(cx.data.1.AllData |>
              filter(year == 2016), by = "series_id") |>
  inner_join(cx.item |>
              select(2, 3), by = c("item_code")) |>
  select(3, 4, 5, item_text, value) |>
  arrange(-value) %>%
  {if (is_html_output()) datatable(., filter = 'top', rownames = F) else .}

Consumption and Saving (Econ 102)

$ Saving

cx.series |>
  filter(demographics_code == "LB15",
         item_code %in% c("INCAFTTX", "HOUSING", "PENSIONS", "TOTALEXP", "OWNMORTG")) |>
  inner_join(cx.data.1.AllData |>
              filter(year == 2016), by = "series_id") |>
  inner_join(cx.characteristics |>
              select(1, 2, 3), by = c("characteristics_code", "demographics_code")) |>
  select(characteristics_text, item_code, value) |>
  spread(item_code, value) |>
  transmute(characteristics_text,
            INCAFTTX,
            `Saving` = INCAFTTX - TOTALEXP,
            `Total Saving (incl. Pensions)` = INCAFTTX - TOTALEXP + PENSIONS) |>
  gather(variable, value, -1, -2) |>
  ggplot() + geom_line(aes(x = INCAFTTX/1000, y = value/1000, color = variable)) + 
  theme_minimal() + xlab("Income after taxes") + ylab("Saving") +
  theme(legend.position = c(0.45, 0.9),
        legend.title = element_blank()) +
  scale_x_continuous(breaks = seq(0, 200, 20),
                     labels = dollar_format(suffix = "K", accuracy = 1)) +
  scale_color_manual(values = viridis(3)[1:2]) +
  scale_y_continuous(breaks = seq(-100, 200, 10),
                     labels = dollar_format(suffix = "K", accuracy = 1))

Saving Rate

cx.series |>
  filter(demographics_code == "LB15",
         item_code %in% c("INCAFTTX", "HOUSING", "PENSIONS", "TOTALEXP", "OWNMORTG")) |>
  inner_join(cx.data.1.AllData |>
              filter(year == 2016), by = "series_id") |>
  inner_join(cx.characteristics |>
              select(1, 2, 3), by = c("characteristics_code", "demographics_code")) |>
  select(characteristics_text, item_code, value) |>
  spread(item_code, value) |>
  transmute(characteristics_text,
            INCAFTTX,
            `Saving Rate` = (INCAFTTX - TOTALEXP)/INCAFTTX,
            `Total Saving Rate (incl. Pensions)` = (INCAFTTX - TOTALEXP + PENSIONS)/INCAFTTX) |>
  gather(variable, value, -1, -2) |>
  ggplot() + geom_line(aes(x = INCAFTTX/1000, y = value, color = variable)) + 
  theme_minimal() + xlab("Income after taxes") + ylab("Saving Rate") +
  theme(legend.position = c(0.65, 0.3),
        legend.title = element_blank()) +
  scale_x_continuous(breaks = seq(0, 200, 20),
                     labels = dollar_format(suffix = "K", accuracy = 1)) +
  scale_color_manual(values = viridis(3)[1:2]) +
  scale_y_continuous(breaks = 0.01*seq(-500, 200, 20),
                     labels = percent_format(accuracy = 1))

$ Saving (Log)

cx.series |>
  filter(demographics_code == "LB15",
         item_code %in% c("INCAFTTX", "HOUSING", "PENSIONS", "TOTALEXP", "OWNMORTG")) |>
  inner_join(cx.data.1.AllData |>
              filter(year == 2016), by = "series_id") |>
  inner_join(cx.characteristics |>
              select(1, 2, 3), by = c("characteristics_code", "demographics_code")) |>
  select(characteristics_text, item_code, value) |>
  spread(item_code, value) |>
  transmute(characteristics_text,
            INCAFTTX,
            `Saving` = INCAFTTX - TOTALEXP,
            `Total Saving (incl. Pensions)` = INCAFTTX - TOTALEXP + PENSIONS) |>
  gather(variable, value, -1, -2) |>
  ggplot() + geom_line(aes(x = INCAFTTX/1000, y = value/1000, color = variable)) + 
  theme_minimal() + xlab("Income after taxes") + ylab("Saving") +
  theme(legend.position = c(0.45, 0.9),
        legend.title = element_blank()) +
  scale_x_log10(breaks = c(10, 20, 40, 80, 100, 160, 200),
                     labels = dollar_format(suffix = "K", accuracy = 1)) +
  scale_color_manual(values = viridis(3)[1:2]) +
  scale_y_continuous(breaks = seq(-100, 200, 10),
                     labels = dollar_format(suffix = "K", accuracy = 1))

Saving Rate (Log)

cx.series |>
  filter(demographics_code == "LB15",
         item_code %in% c("INCAFTTX", "HOUSING", "PENSIONS", "TOTALEXP", "OWNMORTG")) |>
  inner_join(cx.data.1.AllData |>
              filter(year == 2016), by = "series_id") |>
  inner_join(cx.characteristics |>
              select(1, 2, 3), by = c("characteristics_code", "demographics_code")) |>
  select(characteristics_text, item_code, value) |>
  spread(item_code, value) |>
  transmute(characteristics_text,
            INCAFTTX,
            `Saving Rate` = (INCAFTTX - TOTALEXP)/INCAFTTX,
            `Total Saving Rate (incl. Pensions)` = (INCAFTTX - TOTALEXP + PENSIONS)/INCAFTTX) |>
  gather(variable, value, -1, -2) |>
  ggplot() + geom_line(aes(x = INCAFTTX/1000, y = value, color = variable)) + 
  theme_minimal() + xlab("Income after taxes") + ylab("Saving Rate") +
  theme(legend.position = c(0.65, 0.3),
        legend.title = element_blank()) +
  scale_x_log10(breaks = c(10, 20, 40, 80, 100, 160, 200),
                     labels = dollar_format(suffix = "K", accuracy = 1)) +
  scale_color_manual(values = viridis(3)[1:2]) +
  scale_y_continuous(breaks = 0.01*seq(-500, 200, 20),
                     labels = percent_format(accuracy = 1))

Saving Rate (Log) > 30K

cx.series |>
  filter(demographics_code == "LB15",
         item_code %in% c("INCAFTTX", "HOUSING", "PENSIONS", "TOTALEXP", "OWNMORTG")) |>
  inner_join(cx.data.1.AllData |>
              filter(year == 2016), by = "series_id") |>
  inner_join(cx.characteristics |>
              select(1, 2, 3), by = c("characteristics_code", "demographics_code")) |>
  select(characteristics_text, item_code, value) |>
  spread(item_code, value) |>
  transmute(characteristics_text,
            INCAFTTX,
            `Saving Rate` = (INCAFTTX - TOTALEXP)/INCAFTTX,
            `Total Saving Rate (incl. Pensions)` = (INCAFTTX - TOTALEXP + PENSIONS)/INCAFTTX) |>
  gather(variable, value, -1, -2) |>
  filter(INCAFTTX > 30000) |>
  ggplot() + geom_line(aes(x = INCAFTTX/1000, y = value, color = variable)) + 
  theme_minimal() + xlab("Income after taxes") + ylab("Saving Rate") +
  theme(legend.position = c(0.65, 0.3),
        legend.title = element_blank()) +
  scale_x_log10(breaks = c(10, 20, 30, 40, 50, 60, 70, 80, 100, 120, 140, 160, 200),
                     labels = dollar_format(suffix = "K", accuracy = 1)) +
  scale_color_manual(values = viridis(3)[1:2]) +
  scale_y_continuous(breaks = 0.01*seq(-500, 200, 10),
                     labels = percent_format(accuracy = 1))

$ Consumption

cx.series |>
  filter(demographics_code == "LB15",
         item_code %in% c("INCAFTTX", "HOUSING", "PENSIONS", "TOTALEXP", "OWNMORTG")) |>
  inner_join(cx.data.1.AllData |>
              filter(year == 2016), by = "series_id") |>
  inner_join(cx.characteristics |>
              select(1, 2, 3), by = c("characteristics_code", "demographics_code")) |>
  select(characteristics_text, item_code, value) |>
  spread(item_code, value) |>
  transmute(characteristics_text,
            INCAFTTX,
            `Total Expenditures (incl. Pensions)` = TOTALEXP,
            `Total Expenditures` = TOTALEXP - PENSIONS) |>
  gather(variable, value, -1, -2) |>
  ggplot() + geom_line(aes(x = INCAFTTX/1000, y = value/1000, color = variable)) + 
  theme_minimal() + xlab("Income after taxes") + ylab("Consumption") +
  theme(legend.position = c(0.45, 0.9),
        legend.title = element_blank()) +
  scale_x_continuous(breaks = seq(0, 200, 20),
                     labels = dollar_format(suffix = "K", accuracy = 1)) +
  scale_color_manual(values = viridis(3)[1:2]) +
  scale_y_continuous(breaks = seq(-100, 200, 10),
                     labels = dollar_format(suffix = "K", accuracy = 1))

Consumption Rate

cx.series |>
  filter(demographics_code == "LB15",
         item_code %in% c("INCAFTTX", "HOUSING", "PENSIONS", "TOTALEXP", "OWNMORTG")) |>
  inner_join(cx.data.1.AllData |>
              filter(year == 2016), by = "series_id") |>
  inner_join(cx.characteristics |>
              select(1, 2, 3), by = c("characteristics_code", "demographics_code")) |>
  select(characteristics_text, item_code, value) |>
  spread(item_code, value) |>
  transmute(characteristics_text,
            INCAFTTX,
            `Consumption Rate` = (TOTALEXP)/INCAFTTX,
            `Consumption Rate (incl. Pensions)` = (TOTALEXP - PENSIONS)/INCAFTTX) |>
  gather(variable, value, -1, -2) |>
  ggplot() + geom_line(aes(x = INCAFTTX/1000, y = value, color = variable)) + 
  theme_minimal() + xlab("Income after taxes") + ylab("Consumption Rate") +
  theme(legend.position = c(0.65, 0.3),
        legend.title = element_blank()) +
  scale_x_continuous(breaks = seq(0, 200, 20),
                     labels = dollar_format(suffix = "K", accuracy = 1)) +
  scale_color_manual(values = viridis(3)[1:2]) +
  scale_y_continuous(breaks = 0.01*seq(-500, 200, 20),
                     labels = percent_format(accuracy = 1))

$ Consumption (Log)

cx.series |>
  filter(demographics_code == "LB15",
         item_code %in% c("INCAFTTX", "HOUSING", "PENSIONS", "TOTALEXP", "OWNMORTG")) |>
  inner_join(cx.data.1.AllData |>
              filter(year == 2016), by = "series_id") |>
  inner_join(cx.characteristics |>
              select(1, 2, 3), by = c("characteristics_code", "demographics_code")) |>
  select(characteristics_text, item_code, value) |>
  spread(item_code, value) |>
  transmute(characteristics_text,
            INCAFTTX,
            `Total Expenditures (incl. Pensions)` = TOTALEXP,
            `Total Expenditures` = TOTALEXP - PENSIONS) |>
  gather(variable, value, -1, -2) |>
  ggplot() + geom_line(aes(x = INCAFTTX/1000, y = value/1000, color = variable)) + 
  theme_minimal() + xlab("Income after taxes") + ylab("Consumption") +
  theme(legend.position = c(0.3, 0.9),
        legend.title = element_blank()) +
  scale_x_continuous(breaks = seq(0, 200, 20),
                     labels = dollar_format(suffix = "K", accuracy = 1)) +
  scale_color_manual(values = viridis(3)[1:2]) +
  scale_y_log10(breaks = seq(-100, 200, 10),
                     labels = dollar_format(suffix = "K", accuracy = 1))

Consumption Rate (Log)

cx.series |>
  filter(demographics_code == "LB15",
         item_code %in% c("INCAFTTX", "HOUSING", "PENSIONS", "TOTALEXP", "OWNMORTG")) |>
  inner_join(cx.data.1.AllData |>
              filter(year == 2016), by = "series_id") |>
  inner_join(cx.characteristics |>
              select(1, 2, 3), by = c("characteristics_code", "demographics_code")) |>
  select(characteristics_text, item_code, value) |>
  spread(item_code, value) |>
  transmute(characteristics_text,
            INCAFTTX,
            `Consumption Rate` = (TOTALEXP)/INCAFTTX,
            `Consumption Rate (incl. Pensions)` = (TOTALEXP - PENSIONS)/INCAFTTX) |>
  gather(variable, value, -1, -2) |>
  ggplot() + geom_line(aes(x = INCAFTTX/1000, y = value, color = variable)) + 
  theme_minimal() + xlab("Income after taxes") + ylab("Consumption Rate") +
  theme(legend.position = c(0.65, 0.8),
        legend.title = element_blank()) +
  scale_x_log10(breaks = c(10, 20, 40, 80, 100, 160, 200),
                     labels = dollar_format(suffix = "K", accuracy = 1)) +
  scale_color_manual(values = viridis(3)[1:2]) +
  scale_y_continuous(breaks = 0.01*seq(-500, 600, 20),
                     labels = percent_format(accuracy = 1))

Expenditures

Main: Housing, Food, Transportation

cx.series |>
  filter(demographics_code == "LB15",
         item_code %in% c("TOTALEXP", "HOUSING", "FOODTOTL", "TRANS", "INCAFTTX")) |>
  inner_join(cx.data.1.AllData |>
              filter(year == 2016), by = "series_id") |>
  inner_join(cx.characteristics |>
              select(1, 2, 3), by = c("characteristics_code", "demographics_code")) |>
  select(characteristics_text, item_code, value) |>
  spread(item_code, value) |>
  transmute(characteristics_text,
            INCAFTTX,
            `Housing (% of Expenditure)` = HOUSING/TOTALEXP,
            `Food (% of Expenditure)` = FOODTOTL/TOTALEXP,
            `Transportation (% of Expenditure)` = TRANS/TOTALEXP) |>
  gather(variable, value, -1, -2) |>
  filter(INCAFTTX > 20000) |>
  ggplot() + geom_line(aes(x = INCAFTTX/1000, y = value, color = variable)) + 
  theme_minimal() + xlab("Income after taxes") + ylab("% of Expenditure") +
  theme(legend.position = c(0.65, 0.9),
        legend.title = element_blank()) +
  scale_x_continuous(breaks = seq(0, 200, 20),
                     labels = dollar_format(suffix = "K", accuracy = 1)) +
  scale_color_manual(values = viridis(4)[1:3]) +
  scale_y_continuous(breaks = 0.01*seq(-500, 200, 5),
                     labels = percent_format(accuracy = 1),
                     limits = c(0, 0.5))

Housing Rate

cx.series |>
  filter(demographics_code == "LB15",
         item_code %in% c("INCAFTTX", "HOUSING", "PENSIONS", "TOTALEXP", "OWNMORTG")) |>
  inner_join(cx.data.1.AllData |>
              filter(year == 2016), by = "series_id") |>
  inner_join(cx.characteristics |>
              select(1, 2, 3), by = c("characteristics_code", "demographics_code")) |>
  select(characteristics_text, item_code, value) |>
  spread(item_code, value) |>
  transmute(characteristics_text,
            INCAFTTX,
            `Housing (% of Expenditure)` = HOUSING/TOTALEXP,
            `Housing (% of Income)` = HOUSING/INCAFTTX) |>
  gather(variable, value, -1, -2) |>
  filter(INCAFTTX > 20000) |>
  ggplot() + geom_line(aes(x = INCAFTTX/1000, y = value, color = variable)) + 
  theme_minimal() + xlab("Income after taxes") + ylab("Housing Consumption (%)") +
  theme(legend.position = c(0.65, 0.8),
        legend.title = element_blank()) +
  scale_x_continuous(breaks = seq(0, 200, 20),
                     labels = dollar_format(suffix = "K", accuracy = 1)) +
  scale_color_manual(values = viridis(3)[1:2]) +
  scale_y_continuous(breaks = 0.01*seq(-500, 200, 5),
                     labels = percent_format(accuracy = 1))