Social Security on ...
 
Notifications
Clear all

Social Security on Tabular Projections--Income

20 Posts
5 Users
0 Likes
1,590 Views
(@tchamber236)
Active Member Customer
Joined: 2 years ago
Posts: 6
Topic starter  

I'm struggling with two things in the Social Security income stream on the income table. Set to Today$, I would expect it to be a constant dollar amount after the first year. That's how it shows in the manual. Instead, it declines substantially every year until it reaches about half the initial amount. If I put in a COLA tweak--say, -0.1%--in the input table, it behaves as I would expect, reducing the Today$ amount by a tenth of a percent every year. But when the COLA adjustment input is left blank the results make no sense. What am I doing wrong?

Second issue I have is with specifying my social security amount at FRA. I don't know what that is or how to get it. I'm 68, and expect to claim next year at 70. SSA reports the amount I am entitled to today, and the amount I will be entitled to next year at 70. But not my FRA amount. I can understand that since I passed that two years ago and it no longer has any meaning. I reverse engineered an amount and plugged it in, but I wouldn't think I should have to do that. Since I know the amount I will be entitled to next year, isn't there a way to just plug it in?


   
ReplyQuote
(@smatthews51)
Member Admin
Joined: 4 years ago
Posts: 689
 

Dame,

I don't know what you might be doing wrong such that it appears that you're not seeing a constant dollar amount when looking at SS benefits in terms of today's dollars. I verified on multiple versions that PRC is working correctly in this regard. If you'd send me (mail@pralanaconsulting.com) an export file or even a screenshot of the relevant pages I might be able to provide more help.

Your point is well taken about SSA reports not showing your FRA amount after you've passed that age but, no, there isn't a way to "just plug in" the amount you're expecting to receive UNLESS you do it via some other section of the Income page. PRC is designed to help you determine the best age to start benefits, and it does this by modeling all the alternatives based on your FRA amount. Also, PRC models spousal benefits and they're based on FRA amounts, so the tool needs that value to do its thing. If that's not important to you, you can simply enter the desired amounts in one of the Other Income streams.

Stuart


   
ReplyQuote
(@tchamber236)
Active Member Customer
Joined: 2 years ago
Posts: 6
Topic starter  
Posted by: @smatthews51

Dame,

I don't know what you might be doing wrong such that it appears that you're not seeing a constant dollar amount when looking at SS benefits in terms of today's dollars. I verified on multiple versions that PRC is working correctly in this regard. If you'd send me (mail@pralanaconsulting.com) an export file or even a screenshot of the relevant pages I might be able to provide more help.

Your point is well taken about SSA reports not showing your FRA amount after you've passed that age but, no, there isn't a way to "just plug in" the amount you're expecting to receive UNLESS you do it via some other section of the Income page. PRC is designed to help you determine the best age to start benefits, and it does this by modeling all the alternatives based on your FRA amount. Also, PRC models spousal benefits and they're based on FRA amounts, so the tool needs that value to do its thing. If that's not important to you, you can simply enter the desired amounts in one of the Other Income streams.

Stuart

Thanks, I figured out the first item. It happens when Historial Sequence Analysis is active; when that is inactive the social security benefits are consistent. I don't think that activating sequence analysis should affect social security, should it?

The second item isn't critical. In another year I'll be in payment status and I can just enter the payment I'm receiving.


   
ReplyQuote
(@smatthews51)
Member Admin
Joined: 4 years ago
Posts: 689
 

@tchamber236 The historical sequence invokes both historical rates of return and historical inflation so, yes, it does affect Social Security.


   
ReplyQuote
(@tchamber236)
Active Member Customer
Joined: 2 years ago
Posts: 6
Topic starter  

@smatthews51 I left off on this topic in the spring since I was done modeling for the year but I wanted to come back to it before year-end since I don't think the historical sequence analysis is correct. It is showing a much too high probability of running out of money, and the reason is that the social security benefit is being reduced for inflation.

The point of social security is that it is an inflation-adjusted annuity. (I know PRC has user adjustments for those who are skeptical about that assertion but let's stay with the default.) Historical inflation shouldn't have any impact on social security since SS is, by default, inflation-adjusted. It doesn't matter what may have happened in years past, SS will keep up.

See my first posting above. I can make this error go away by putting in a tiny COLA adjustment, and then PRC will ignore historical inflation and behave correctly. Without it, it gives a seriously wrong answer.

This post was modified 1 year ago by Dame Wang

   
ReplyQuote
(@smatthews51)
Member Admin
Joined: 4 years ago
Posts: 689
 

@tchamber236 Hi Dame,

Let's examine your statement: "Historical inflation shouldn't have any impact on social security since SS is, by default, inflation-adjusted. It doesn't matter what may have happened in years past, SS will keep up." On the contrary, in the historical sequence analysis, what happened in the past is ALL that matters. That's the whole point of the exercise:to project what would happen in your particular scenario IF the specified historical inflation and rate of return sequence were to be repeated. SS benefits are adjusted per inflation, so if inflation goes down, so will your SS benefits. With that said, I took a look at what I think is the worst historical sequence: one that begins in 1965. In that case, the cumulative inflation over a 20-year period would be over 300%, whereas a sequence that began in 1930 would have less than half that, and that results in huge differences in the long term projection of the test case I used.

Regarding your point about being able to make the "error" go away by putting in a tiny COLA adjustment which then causes PRC to ignore historical inflation, I'm curious to know how "tiny" your COLA adjustment was. It doesn't cause PRC to ignore anything but it does skew the whole inflation profile by the specified amount and that can make a big difference over a period of years. I observed substantial movement of the overall projection by adding in a percent or two of COLA to a scenario where SS was the only source of income but I wouldn't call that "tiny".

Stuart


   
ReplyQuote
(@pizzaman)
Honorable Member Customer
Joined: 3 years ago
Posts: 430
 

Warning, Pizzaman is about to go into the weeds, again 😲. Inflation is not the same as Consumer Price Index (CPI). https://www.usinflationcalculator.com/inflation/inflation-vs-consumer-price-index-cpi-how-they-are-different/ CPI comes in many flavors, the two most reported in the media are; the Consumer Price Index for All Urban Consumers (CPI-U), and the Consumer Price Index for Urban Wage Earners and Clerical Workers (CPI-W): https://www.investopedia.com/ask/answers/012115/consumer-price-index-cpi-best-measure-inflation.asp

https://www.bls.gov/opub/btn/volume-3/why-does-bls-provide-both-the-cpi-w-and-cpi-u.htm

As a result, since 1985, the two indexes have differed only in the expenditure weights assigned to item categories and geographic areas. While the CPI-W is used to calculate Social Security (SS) cost-of-living adjustments, most other COLAs cited in federal legislation, such as the indexation of federal income tax brackets, uses the CPI-U.

So, SS's annual benefit increase (COLA) are are not adjusted per inflation, but by the CPI-W. And, COLA can never be negative, even in times of disinflation.

@smatthews51 Page 67 of the PRC manual states: Social Security Administration will increase benefits each year at the rate of general inflation and PRC assumes that as its default approach. If you use PRC's historical sequence analysis which uses historical inflation to determine COLA, and not CPI-W, won't that give you the wrong result?


   
ReplyQuote
(@pizzaman)
Honorable Member Customer
Joined: 3 years ago
Posts: 430
 

Which CPI index is best for determining SS COLA? CPI-E according to many: https://www.ncpssm.org/documents/social-security-policy-papers/the-cpi-e-a-better-option-for-calculating-social-security-colas/


   
ReplyQuote
(@tchamber236)
Active Member Customer
Joined: 2 years ago
Posts: 6
Topic starter  

@smatthews51 Social security is inflation-adjusted. Historical sequences are irrelevant for such an income stream. Whatever your benefit is today is what it will be, in real dollars, until the day you die. That's the default assumption in PRC and the way it works in every other scenario in PRC. Why are you doing it differently here? What inflation was in 1930 or 1965 has no relevance for a 2022 retiree, as regards social security. (Of course it will have an effect on other non-indexed income streams.)

I recognize that many people may disagree that social security is truly "real income for life," but PRC has the flexibility for a user to express their disagreement in the inputs. I imagine many users do make such adjustments somewhere in their scenarios. I know I do. But that's their call.


   
ReplyQuote
(@smatthews51)
Member Admin
Joined: 4 years ago
Posts: 689
 

@tchamber236 I need to make a correction to my response; I answered too quickly and without fully remembering how the historical analysis actually works. So, during a historical analysis, income streams are NOT modified in any way (only the expense streams and rates of return are effected by the historical record). If you've set your SS income stream to be, say, $40,000, then it will stay at that level adjusted annually at the inflation rate specified on the Home page + whatever delta you choose to add in the SS section of the Income page. So, if you look at the income projections in terms of future dollars, you'll see that the SS income steps up every year at the specified rate. On the other hand, if you look at it in terms of today's dollars, you may see it go up or down, depending on the historical inflation associated with the selected historical sequence. In a sequence with above-average inflation over time, your SS income will have decreased buying power and that will be reflected as decreasing SS income over time.

Stuart


   
ReplyQuote
(@smatthews51)
Member Admin
Joined: 4 years ago
Posts: 689
 

@pizzaman Yes, of course, you're right about SS being adjusted based on the CPI. When I read your post it immediately occurred to me that my response was incorrect, so I added a follow-on correction. The fact is that PRC doesn't make ANY adjustments to any income streams during an historical analysis or historical sequence analysis; in particular, SS income continues to escalate as it does in a fixed rate analysis. The reason for this is that most income streams aren't necessarily tied to the inflation rate or the CPI, so I have no basis for making adjustments based on the historical record. This is all explained in the manual; I simply overlooked that fact in making my initial (hasty) response. Thanks for chiming in!

Stuart


   
ReplyQuote
(@smatthews51)
Member Admin
Joined: 4 years ago
Posts: 689
 

@tchamber236 Please note that I made a correction to my initial response. Pizzaman's post caused me to rethink this and I then realized this fact: PRC's historical analysis doesn't make ANY adjustments to the SS income stream or any other income stream. It continues escalating at the specified rate (the same one used in the fixed rate analysis) regardless of historical inflation rates. You can confirm this by looking at the income projection in future dollars. With that said, looking at the income projection in terms of today's dollars can be a different story, depending on which historical sequence is selected. If the selected stream has higher long term inflation than the fixed inflation rate projection, then the income streams will decrease over time.

I recognize that this introduces some error into the analysis but there's no alternative in the current design because not all income streams are tied to the inflation rate. I'm putting this on the list of potential enhancements for the future but success would entail making some decent assumptions about the correlation between ALL income streams and inflation.

Stuart


   
ReplyQuote
(@tchamber236)
Active Member Customer
Joined: 2 years ago
Posts: 6
Topic starter  

@smatthews51 What I am seeing is this: in "Today" dollars, my social security is exactly the same amount through the end of the model. Never changes. That is in both historical and Monte Carlo analyses. That is what I would expect--it's the premise of social security.

When I click the "sequence" button, the social security amount in Today's dollars falls off a cliff, ending at half the beginning amount. This is not what should happen. It's wrong, and materially wrong. So wrong as to make "historical sequence analysis" a useless measure for anyone depending on social security to contribute significantly to their retirement. The "sequence" button shouldn't change the already computed social security amounts.

I don't want to address indexed income streams as a class because I don't know how commonplace they are. Social security is a different animal. It's ubiquitous. It should work correctly across the board because everyone has it, and it is likely to be a significant income stream for most people.


   
ReplyQuote
(@pizzaman)
Honorable Member Customer
Joined: 3 years ago
Posts: 430
 

Going into the weeds update: CPI-W has been in existence since 1974. https://www.ssa.gov/oact/STATS/cpiw.html Taking the value of CPI-W on December of each year from 1974 to 2021 and averaging the annual change, you get 3.64%. If you take the general inflation number from PRC Historic Data tab for the same years, you get an average of 3.66%. The corresponding individual years can very somewhat, but over time it appears to average out. So.....going into the weeds resulted in... not much 😫 O'well.


   
ReplyQuote
(@smatthews51)
Member Admin
Joined: 4 years ago
Posts: 689
 

@tchamber236 You may be happy to know that the 2023 version of PRC (due to be released in January 2023) has a new implementation for modeling inflation and it will yield two major advantages over all prior models: 1) for fixed rate and Monte Carlo analyses, the user can specify different inflation rates for each of two separate time periods and 2) it will enable income streams tied to inflation (such as Social Security) to be modeled correctly in historical analyses where the inflation rate isn't constant. Depending on the starting year of the selected historical sequence, Social Security income will vary in terms of future dollars but it will be a constant value in terms of today's dollars regardless of the historical sequence.


   
ReplyQuote
Page 1 / 2
Share: