ACS Summary file for ZCTAs

Not like the summary files, but the summary files themselves have ZCTAs: https://www2.census.gov/programs-surveys/acs/summary_file/2021/sequence-based-SF/data/5_year_seq_by_state/UnitedStates/All_Geographies_Not_Tracts_Block_Groups/. They're only in the UnitedStates files, not the state files, because ZCTAs cross state boundaries. I'm not sure if this is what you're referring to with summary files; they're a little unwieldy (the data files don't have headers, if I recall, and you have to look up which sequence table has the ACS table you want).

Not like the summary files, but the summary files themselves have ZCTAs: https://www2.census.gov/programs-surveys/acs/summary_file/2021/sequence-based-SF/data/5_year_seq_by_state/UnitedStates/All_Geographies_Not_Tracts_Block_Groups/. They're only in the UnitedStates files, not the state files, because ZCTAs cross state boundaries. I'm not sure if this is what you're referring to with summary files; they're a little unwieldy (the data files don't have headers, if I recall, and you have to look up which sequence table has the ACS table you want).

Thanks Bernie !

A few minutes after I posted I figured out the  geo type for ZCTA (860)..  I was reading the the documentation and they only gave a few examples and ZCTA wasn't listed.   The summary files are pipe "|" delimited with headers.  You have to pull out the right column.  I was looking for total population  B01001_E001.  

I'm trying to merge zip codes so the the combined area has a population of 20,000 or more.  I'm trying to de-identify a data set that has zip codes by merging adjacent ZCTAs.

If you have any ideas let me know.   Perhaps someone has already done this and they have a list.

ZTCA is a zip-similar that approximates a Zip Codes. Zip's themselves are a bundle of postal delivery routes of which 11% or so are updated every 60 days or so. They are optimized for mail delivery, not statistical purposes. And there are no Zip-ZTCA crosswalk files available commercially that I know of. 

Tracts average 3,060 voting  age persons which will be easier to match into 20k regions, Block Groups are the next-best option at 1,880 persons compared to a list of zipcodes referenced on Jan 1 of this year, at 8,230 per general delivery Zipcode (excluding PO Boxes and dedicated or delivery-point codes).

Now, I'm saying this with no knowledge of your application. If it is important that you need known zip areas (90210, for example) then you are stuck with Zips or ZCTA and hope you can stich the two regions together. 

My opinion; you're going to need to decide on zip, or zip-similar but not both. If your objective is to package a set of boundaries to precise 20k populations then I recommend using something more granulay, like Census Tracts or Block groups..The other advantage from using these smaller regions is that they strictly conform to state and county boundaries.