I am hoping to express ACS 2013-2017 dollar-denominated data as 2010 dollars. Following this resource, my understanding is that I would take the average CPI value for 2010 and divide it by the average CPI value of 2017 (the most recent/last year in the series). This would give me the inflation adjustment ratio which I would then multiply by an ACS estimate for 2013-2017. Why is the CPI value for the last year in the series used (i.e. 2017) rather than an average CPI value for all 5-years (i.e. 2013-2017)?

Parents
• Census provides specific advise for comparing ACS data.  For the 2017 ACS release look here:

https://www.census.gov/programs-surveys/acs/guidance/comparing-acs-data.2017.html

In sum, the recommendation is that one use the CPI Research Series to conduct comparisons where you inflate or deflate reported values.  You can find  information about the values for the CPI-U-RS (or R-CPI-U-RS as it now seems to be called) here:

https://www.bls.gov/cpi/research-series/r-cpi-u-rs-home.htm

My recollection, and someone please correct me if wrong, is that ACS dollar values are adjusted to the last month in the period, so to adjust 2013-17 values down to the average value for 2010 you would use the ratio of the 2010 average to the December 2017 index figure.

• Thanks, Cliff. Based on your note about  ACS dollars being adjusted to the last month in the period, I looked into this a bit more to see how multi-year estimates are created. I found this Census resource (page 22) which explains how multiyear estimates are created and provides an example which is helpful. Here is the excerpt (emphasis added):

"Creating Values Used in Multiyear Estimates
Multiyear income, rent, home value, and energy cost values are created with inflation adjustments. The Census Bureau uses the All Items CPI-U-RS Annual Averages for each year in the multiyear time period to calculate a set of inflation adjustment factors. Adjustment factors for a time period are calculated as ratios of the CPI-U-RS Annual Average from its most recent year to the CPI-U-RS Annual Averages from each of its earlier years. The ACS values for each of the earlier years in the multiyear period are multiplied by the appropriate inflation adjustment factors to produce the inflation-adjusted values. These values are then used to create the multiyear estimates.
As an illustration, consider the time period 2004–2006, which consisted of individual reference-year income values of \$30,000 for 2006, \$20,000 for 2005, and \$10,000 for 2004. The multiyear income components are created from inflation-adjusted reference period income values using factors based on the All Items CPI-U-RS Annual Averages of 277.4 (for 2004), 286.7 (for 2005), and 296.1 (for 2006). The adjusted 2005 value is the ratio of 296.1 to 286.7 applied to \$20,000, which equals \$20,656. Similarly, the 2004 value is the ratio of 296.1 to 277.4 applied to \$10,000, which equals \$10,674."
Within a ACS series it looks like everything is adjusted to the CPI annual average of the last year in order to generate the multi-year estimate. Therefore, the dollars are expressed as dollars in the last year--in the excerpt above, the dollars for 2004-2006 are expressed as 2006 dollars. Given this, I now understand why the ACS resource I linked in my original post directs users to use the CPI annual average of the last year in the series when adjusting/deflating the measures to an earlier year.