It says something about the glamorous life I lead that my heart skipped a beat last night when I went through the daily mail and found the census form had arrived. "Oh cool, the census," I actually said as my wife leaned in for acknowledgment that I was glad to see her, too.
Only it wasn't the census form. It was a letter saying the census form would be arriving in about a week.
Want to guess how much this little exercise cost? Hand me that napkin.
There were 105,480,101 households in 2000. At 500 sheets of paper per ream, that's 210,960 reams of paper for the letter. It's cheap paper, though. At $40 a case, Office Max has the cheapest price I could find online, so that's $843,000 for the paper.
Five-hundred envelopes go for $30. That's another $6.3 million (I'm rounding up and down here; it's the government afterall).
Finally, there's the cost of mailing. It's presorted first-class mail. According to the U.S. Postal Service Web site, pre-sorted mail costs .335, although a standard rate letter could be sent for 17 cents. But this was first-class. Total: $35,335,833.83.
Total: $42.5 million (although I remain somewhat skeptical about the postage) to send you a letter to tell you you're going to get another letter next week. Oh, and sending a postcard would've been $15.8 million cheaper.
The average person pays $13,000 in federal taxes per year. So it took the annual federal taxes of nearly 327 taxpayers to send you the letter.
(See "Why Use Advance Letters")
Ha! I'm glad I'm not the only one who thought that a receiving a letter telling me that I am going to receive a letter is a little silly to say the least.
And some people wonder why others don't want government involved in health care. Not that the census is run by the same people, because goverment will hire the smart people to run health care.
I had this same conversation with my wife last night when we got our letter. Add on the ridiculous "ad" campaign that went along with the census and it seems to me that the government spent gobs of money unnecessarily.
$42.5 Million just in paper, postage, and envelopes. Now factor in the thousands of personnel hours that went into preparing the mailing, all at a government union wage, probably in the line of $20-$25 an hour or more. It's okay, just add it to the deficit and the national debt, someone will pay for it eventually.
This post is a perfect example of why I want to gouge my eyes out any time a journalist proudly boasts that they became a reporter because they can't do math.
I got home late last night and just put that aside figuring I'd deal with it later in the week. Now I don't have to.
I'll see your $35 million and raise about twice that at least. Usually the actual questionnaire is many pages of heavier stock paper and sent as a "flat" (8.5 x 11 envelope.) The cost of mailing that will easily be twice that of a standard #10 business envelope with one sheet of paper. Some of the addresses may no longer be residences or even exist. So if you want to make sure that the form packet can be delivered the most accurate way to do it is to send out something and see what "bounces back". You let the post office confirm the addresses for you. The cost of printing the letters is probably less than printing a post card to the point that it makes up for the difference in postage. Also dealing with the bounced letters is easier than the packets if you want to send a census worker to the site and verify. Also plain paper is easier than postcards. Have you ever tried to deal with a stack of post cards on a clipboard?
I don't know that these are the reasons for the letter, but it they make sense to me.
This is standard practice in the survey field when done by private business -- And they do it because it works to increase participation. Given that the government has to spend big $ to send enumerators out to households that fail to respond to census mailings, I would think we'd all be happy to spend a little now if it saves a lot tomorrow.
The self-selected sample of geeks who read Bob's blog (all of whom are probably not only anxious to get their census forms, but secretly longing to get the long form) is hardly representative of the full population. So we should all chill and maybe on occasion take a moment to think that the government might actually be doing something smart.
Figure the cost of this:
1. In November I received the 2010 Census form, filled it out and sent it back.
2. In December, someone from the Census called me to confirm some information.
3. In February I received another letter telling me I would receive the Census form by mail soon. (Remember: I'd already filled it out and sent it back.)
4. Last weekend, a census worker came to my home, and personally delivered another census form. He stated I HAD to take from him even though I had already filled out and returned one. I threw it in the garbage.
How much did it costs to get my Census information? What a waste!
I KNOW! I was so disappointed! it wasn't the long form, and then it wasn't even the short form.
At first when I got my letter at home, I thought what a waste. But I do see some of the reasons why this initial letter may help with the participation rate when the "real" form comes out.
It also helps me understand why the 2010 Census is estimated to cost $14 billion. That's about $45 per person (based on 2010 population estimate).
The cost of the Census has been doubling each time for the past several counts.
Source - I heard it initially cited in a GAO report, but this site has a nice little table:
PS - I too was disappointed I have to wait another week! I've been wondering when the survey would come - I guess I have the answer now.
We've gotten two calls from the Census department.
Additionally, a few months ago they sent an agent in a fancy car to ask about our house. There is a full apt in the basement and they wanted to know if if we treated it as a separate address, no we do not.
Last night we received two census letters, I assume one was for the second residence which we do not actually have.
What a waste.
I hope I get points for trying, but my link didn't work.
Here's the website for my $14billion cost estimate: http://www.genealogybranches.com/censuscosts.html
Here is the official Census Bureau reason:
Those comparing this to healthcare reform shouldn't be so smug. How many "Not a Bill" statements have we all received from our efficient, non-government-run healthcare providers?
I echo the comments above. I thought it was a waste, too, when that letter came (and when I realized it wasn't the real census--this is the first year I get to do one!!!)
However, I can appreciate the comments and insight that explain there might be science and established methods behind the publicity campaign. I wonder, though, what has changed to make these sorts of measures necessary to a growing number of citizens that fails to understand the history and purpose of the census. I would therefore argue that Census Bureau is merely reacting to reality, not making a concerted effort to waste money.
There exists an enormous research literature on how to increase the number of responses to surveys. Many of these studies take the following approach: (1) Have two (or more) sets of identical surveys. (2) Send out some surveys in one way (for example, by sending a letter ahead of time informing the respondent that the survey is coming). (3) Send out another set of surveys in another way (for example, by sending a postcard instead of a letter, or by having no pre-notification at all). (4) Compare the returned surveys and discover which method produced the highest response rate. To see what this research says about pre-notification letters, type "response rate" and "pre-notification" into Google and see what pops up.
The Census Bureau is charged with a Sisyphean task -- to get a response from every household. In order for this to be done as efficiently as possible and at the least cost, I certainly want the Bureau to use the latest and best cutting-edge research on what techniques are likely to maximize the response rate on the mailed Census form. Remember, for every household that does not respond by mail, that household will have to be contacted by a Census worker. The cost for a Census worker to contact a household directly is going to be much, much higher than the cost of sending the household a few extra pieces of mail.
One of the country's leading experts on survey response rates is Robert Groves, formerly of the University of Michigan. When Public Opinion Quarterly did a special issue on survey nonresponse in 2006, it was Groves who wrote the opening article summarizing the decades of research on this topic. His current job is Director of the U.S. Census Bureau -- one of the best appointments Obama has made.
//Given that the government has to spend big $ to send enumerators out to households that fail to respond to census mailings, I would think we'd all be happy to spend a little now if it saves a lot tomorrow.
Maybe, but haven't all of those enumerators already been hired.
From Zach's link:
For every 1 percent increase in households that respond by mail, taxpayers save about $85 million in operational costs associated with census takers going door to door to follow up with households that did not mail back the form.
Let's do the math on that one, then.
First, is there some indication of by how much the letter increases participation? At the rate indicated, a 1 percent increase is 1,054,801 households. An $85 million dollars savings means it costs the Census Bureau $80.58 to get one household to answer 10 questions.
I find that hard to believe. First, the census takers are already in the neighborhoods. And they've already been hired. They make between $10 and $20 an hour.
The Census Bureau has already said it takes 10 minutes to answer 10 questions. So let's say it takes 5 minutes to walk up the driveway and 5 minutes to walk back down... or 20 minutes.
That's charitable, I know. So at $10 an hour, it costs $3.33 to get the answers from one household.
What's the other $76.67 going for?
The Analyst: Are you "The Analyst" from the old blog "The Analyst."? Boy, I miss that blog. Does the Census keep track of households who respond? Can it identify those households most likely NOT to respond?
But beyond that, do people REALLY even read these letters and do they retain that information for a week?
Tif W, are you sure you didn't get the American Community Survey in November?
Those comparing this to healthcare reform shouldn't be so smug. How many "Not a Bill" statements have we all received from our efficient, non-government-run healthcare providers? - Donovan
You do know that those "Not a bill" statements are statements of benefits which are required by law, right?
//How many "Not a Bill" statements have we all received from our efficient, non-government-run healthcare providers?
I went over mine with a fine-toothed comb the other day. It came from Medica. All of the providers were X'd out "to protect privacy."
So basically I had 6 pages of submissions from health care providers who were kept anonymous, for services that weren't detailed, and somehow I'm supposed to figure something out from that?
Bob, could you rethink future postings of “back of the envelope calculations”? Making assumptions and putting math to those assumptions doesn’t create fact.
The letter makes no apparent sense, however Michael offers one possible explanation. The appliance version of “cash-for-clunkers” made no apparent sense, however I came up with an explanation for what was being attempted.
If it’s waste, talk to the agencies debunk the explanations and then report it. If it’s true waste, find the true costs (they may be higher) rather than making assumptions.
PS - doubling costs every 10 years is a little high, but not abnormally higher than inflation.
There is no long form this year.
"That's charitable, I know. So at $10 an hour, it costs $3.33 to get the answers from one household.
What's the other $76.67 going for?"
I work in the survey industry. When surveying by phone, it takes about 10 calls to get one complete. You have to not only call when people are home, but you have to find them willing to drop what they're doing and talk to you. The other $76.67 likely goes to followup visits and other administrative costs. If you're a US Census planner, what time of day are you going to send your employees to people's homes to maximize response rates & reduce costs? Sure, there are lots of homes where there is an adult home all day who can take the survey. But there are also many homes where people are only home after dark. Or if they're home, they're alone & trying to keep the 3 yr old and 1 yr old happy & don't really have a bunch of time to answer surveys.
In short, its easy to make some assumptions about how wasteful the program is, while ignoring the reality that the census bureau faces in fulfilling their mandate.
"The Analyst" makes a very strong point, there is decades of very good research that goes into the method of conducting the Census, which is a glorified survey. I think the bigger issue though is that there is no incentive for the commerce department to save money or be more efficient, only to get the results. If a private company were contracted to do the Census, you can be sure I wouldn't have gotten a first class letter last night, it would have been as Bob suggested a postcard. The mailing may be needed as the research suggests, but that doesn't mean money can't be saved in the process.
Everyone is missing the main issue here - the ridiculous idea that we need to do a census (meaning a count of each and everyone one of our citizens) as opposed to doing a survey of a random sample of citizens. The later would be much more cost effective and also produce a more accurate count of the actual number of citizens.
But there's this little thing called the constitution standing in the way of doing a survey instead of the census. Oh, and that most Americans wouldn't ever believe that a survey could be more accurate than a census.
I know you don't like it, Tired Boomer, but I reserve the right to calculate how much things cost.
I believe doubling costs every 10 years represents a 10% inflation rate. That is abnormally higher than inflation.
Take the last ten years. Something that cost $1 in 2000, would cost $1.28 today. I would argue that $2 IS abnormally higher than $1.28.
As for the appliance program, I saw your assertion and I already understood the motive behind it, but that doesn't mean it made sense. It didn't. There's a big difference between explaining a motive and determining a result.
be sure I wouldn't have gotten a first class letter last night, it would have been as Bob suggested a postcard. The mailing may be needed as the research suggests, but that doesn't mean money can't be saved in the process.
Bingo! Also, if the letter had been sent standard rate, it would have cost half of what it cost.
wondery why they dont just use email instead since they already have access to our email accts at google, yahoo, aol, etc. it would cheaper!
From the Census Bureau:
“The Census Bureau will spend about $25 per person if we have to go out and knock on the doors of households that don’t mail them back.”
Funny how many of us had this same conversation last night with ourselves or our significant others! My fiance told me to write "Look forward to it!" on the letter and mail it back to them. I got a good laugh out of that but decided against doing so.
I just interviewed a guy from the Census on this very Good Question (for tonight's 10pm news on WCCO). He claims that the pre-letter increases responses by 6%. Which makes it a huge money saver.
I asked myself the same question last night. How much did this cost?
I'm glad someone took the time to figure it out. What a waste.
I've taught graduate level marketing, and what the Census did was designed to increase participation and reduce costs.
The reporter should have contacted Census about this or called someone knowledgeable about surveys.
Even if you don't buy the $85 million/1 percent savings number, you only have to save $40 million to make this money worth spending, right?
The census should be outsourced!!!
Poor reporttting? How about your poor spelling!
Don't care if you taught marketing-- waste is waste is waste. In these times, every census penny counts.
"From the Census Bureau:
“The Census Bureau will spend about $25 per person if we have to go out and knock on the doors of households that don’t mail them back.”"
Per person or per address? If the former, that starts coming close to the $80/household earlier, doesn't it?
How about a lottery drawing for the winner to appear on a reality show. That should bring in a few more census completions!
Have to get back to work-- have to earn to pay my taxes to get census letters out.
//He claims that the pre-letter increases responses by 6%. Which makes it a huge money saver.
The Census math is most interesting.
We know -- because the Census people told us -- that 72 percent of 120 million households send the form back.. That's 86 million households.
Without the letter, then 79 million would've sent it back.
The numbers in the original post were based on 105 million, but the Census Bureau says 120 million households are involved here, but about 12 million forms are delivered by hand. I don't know if people who are getting their forms delivered by hand get a letter ahead of time.
So let's go with 108 million. The cost of postage and paper is $43.3 million.
That's $6.19 per every household to get the 7 million people to send theirs back.
But that still leaves 29 million people who haven't send theirs back.
The Census Bureau says $85 million is saved for every 1 percent. So that means that by investing $43.5 million in the pre-letter, it's saving $595 million.
I'm thinking that if you can save $552 million with one letter, why not send two and save a billion? (g)
Actually, Bob and CHS_
a private firm WOULD send a first class letter, b/c as An Analyst says, there is a tremendous body of research showing what kinds of contacts matter to improve response rates. Postcards are for after-questionnaire reminders.
Don Dillman's book is the premiere survey research guide. I for one am glad to see the US Census Bureau is using these proven methods to improve response.
Agreed. But here's a question for you Bob, oh ye of back of the envelope calculations:
If that letter increases the cooperation yield at all, it will save the Census money in funds not paid to Census takers in April.
So the Question is, what percent increase in the response rate does the letter need to achieve in order to pay for itself?
The census costs an average of $45/person. (via http://www.genealogybranches.com/censuscosts.html)
$42.5 million / $45 a person is about 1,000,000.
So if a million extra people turn in the census than otherwise would have, the letter pays for itself.
That's equivalent to about a .33% increase in the response rate.
Worth it? Maybe.
(PS, it was nice to meet you Friday at TPT)
Yes, they have already hired the enumerators (or, more accurately, determined how many they think they'll need, and will soon have them hired).
But, we might also presume that when they budgeted for how many of them to hire they factored in their presumptions of likely response rates to the mailed forms, and that those presumptions included their sense of how much benefit they'd get from this pre-mailing. So, in that sense, we've already received the benefits of the pre-mailing, since they've likely hired fewer of them than they might otherwise have done.
(The '$25 per household' figure is misleading, in that it's based on the presumptions of how many enumerators are hired divided up by the number of households they'll likely need to enumerate. It's not like they get a bill for $25 every time another household gets enumerated -- Most of that is fixed costs that won't change unless the numbers are drastically wrong.)
Again -- Probably not waste (and happy to see others backing this up!).
//If that letter increases the cooperation yield at all, it will save the Census money in funds not paid to Census takers in April.
Right. That's where the math dispute is. It costs $6 to get a person who otherwise wouldn't, to send the Census Form back.
The Census Bureau claims if they don't, it'll cost $85 million to get the Census forms back from 1.2 million people - or $70.83 per household.
Now, sure... $6 compared to $70.83 is a bargain. That's certainly one way of looking at it. The other is $70.83 is a lot of money to spend to send a $10 an hour census worker to someone's house.
The Census Bureau hired 750,000 people "most of whom will work on non-response" according to the Hill.
How many more households would each of those people have to see if the "$40 million letter" hadn't been spent. I get 9.3. over the course of the expected two months of their employment.
One other question for the survey experts who've posted: Is the typical response increase being cited a response rate for a mailed survey or for a phone call survey. I'm aware that the head of the Census is said to have "written the book" on this subject, but it talks about his work on the effect of an advance letter on a telephone survey.
In the interest of accuracy, there is what is known as the rule of 72 in calculating the amount of increase by compound interest.
That rule indicates that doubling an amount in 10 years would be 7.2%, not the 10% you assumed. 10% compounded over 8 years is 214.358881% of the original amount.
I'm not I'm assuming anything other than math.
If something cost $1 in 2000 and cost $2 today, isn't the amount of increase 100% over that time?
I don't know how to make it any other number.
Just breezing through Google on this topic and, umm, wow!
I'm pretty sure the dust-up over the letter will do more to publicize the Census than the letter.
Now I have to calculate how much one blog post will save the U.S. government.
Bob, you are correct. I checked out the website and the two surveys are connected to one another. See link: http://www.census.gov/acs/www/ I do remember clearly that it was the Census office that called me to clarify information, which is what led to my mistake. Sorry!
\\I'm pretty sure the dust-up over the letter will do more to publicize the Census than the letter.
And maybe that was the point. Like the $100 giant flat panel on Black Friday when the store has only 6. Every TV camera is out front covering the event, and that's the real value of it.
Actually, Bob and CHS_
a private firm WOULD send a first class letter, b/c as An Analyst says, there is a tremendous body of research showing what kinds of contacts matter to improve response rates. Postcards are for after-questionnaire reminders.
Don Dillman's book is the premiere survey research guide. I for one am glad to see the US Census Bureau is using these proven methods to improve response.
Posted by Dr_nadine | March 9, 2010 1:37 PM
I agree to a point. In a classical survey example the rate of response is influenced by the rate of contact and the rate of refusal. Pre-notifications are used to increase that contact variable. Things like First Class vs cheaper bulk mailings do make a difference because of the perception of importance associated with the manner of delivery. If something seems more important or more official, you're more likely to respond. However both of these variables are out of whack when you talk about the Census. The vast majority of people know what the Census is, and thus are not going to be influenced by a first class mailing vs standard. Also, the Census is obligatory. Spending money to mail a pre-notification first class seems like a poor use of tax-payer dollars examined in that light.
I do agree that using proven methods and having a well published and respected survey researcher as the head of the census is a good idea, and will likely save us money overall. I just believe that we could be saving more while still getting the results.
I got home last night, it was raining, my Census was in a very nice plastic bag on my door handle. I guess I won a door prize as the bag is now serving as a garbage vessel in my car. Wonder what that cost?
We got the long form in the mail at least a month ago and sent it in promptly. Today when I got home, there was a small plastic bag with a new form in it hanging on the door knob. So if we would fill it in, would we be counted twice?
If you got the long form, it was probably the American Community Survey, an ongoing survey from the Census Bureau. That's the one that Rep. Michele Bachmann confused with the Census when she made it a big issue a few months ago.
It would be great if the reason that the letter was sent was so that bounced letters would be used to prevent the census letters from being printed and mailed.
However, that's not why.
First, these letters were to RESIDENT. At least mine was. Going to RESIDENT, even if no one lives there, the letter doesn't bounce.
Second, the census is being sent NOW. "About one week from now, you will receive a 2010 Census form...". There's NO WAY that the post office could bounce the letters in time for the census department to remove them from the printing and mailing. The census forms have to be start printing within days of you getting the notice letter.
It would take weeks to allow for bounced letters to be processed, not days.
Yes, this is the same person that wrote the old blog. I miss the blog too, but there were good reasons for not continuing it. If you're interested, Bob, we can have lunch sometime and I'll tell you about them.
One other point to add to the discussion -- there is a very active community of researchers and academics that rely heavily on census data (including, for example, the Minnesota State Demographer and the world-renowned Minnesota Population Center at the U of M), and those folks watchdog proposed census changes because of concerns about how their work might be affected. For example, there was tremendous discussion about the decision before the 2000 census to allow respondents to classify themselves as belonging to more than one race. It is simply not the case that the Census Bureau does all its work in secret and then suddenly trots out stupid ideas that would have been quickly debunked had they just let a couple of smart bloggers spend 5 minutes thinking about them first.
I don't think the Census Bureau is infallible - I think it initially headed in the wrong direction on the multiracial issue (it was going to have a single "other" category) - but in this case, I think it's unfortunate that the Bureau is being castigated for following widely accepted and empirically-supported procedures for increasing response rates. That is the nature of the new media environment, though -- the blogosphere and the instant news cycle unfortunately place a premium on those that are fastest on the draw and loudest in a crowd, not on those who actually know something about what they are talking about.
FYI, I have placed two more messages with the Census Bureau for a break down of how they calculate that it costs $75 to send a census worker to each household. I have not heard anything back yet., however
We'll kick this around on the every-Wednesday chat on Morning Edition tomorrow.
A simple 'heads-up" letter with a teaching moment is money well spent, even if the ROI is only one-tenth of what is projected.
If there wasn't the wing-nut ranting that the census is some sort of government conspiracy, perhaps the on-going promotional campaign could have been dialed down.
Perhaps if each census form came with a $1,000,000 Lotto number that is only eligible to win if the census survey is returned and completely filled out, the government could improve its return rate for a lot less money...
That's the ironic part of this. As a result of the fear-mongering by a congresswoman who got the Census confused with the American Community Survey, the government has to spend more money. So we have to spent millions of millions of dollars because of the misinformation campaign by people who are constantly railing against government spending.
Is it a waste of money? Of course it's a waste of money, but perhaps not for the reasons most people think. It's a waste of money because we have to spend $42 million dollars to get people who are too lazy, to fill out the Census. Presumably, this includes the younger demographic which, according to a Pew study, wasn't much interested in the Census.
But I was thinking on the way home about the lotto idea, too. But why $1 million? I wonder what the response rate effect of a prize is. If, as claimed, a half BILLION dollars is saved by this campaign, then what would the effect be of sending out the Census form and having a $20 million jackpot or 20 $1 million jackpots.?
If that raises the response rate by 6 percent, you've just achieved your goals for half the cost.
We're trillions of dollars in debt. We're not going to find it in the couch cushions. As a country, we have no choice but to try to figure out how to do what we're doing now by spending less. Can we get a 6% increase in returns and spend less by, say, using postcards, or sending them out 10 days earlier and sending them standard rate?
Maybe saving $15 million isn't a big deal anymore.
>> The waste of money is the constitutional mandate to have to count every last person. Every statistician knows that a well done survey conducted on a sample of citizens would be more accurate at a much, much lower cost.
This is totally true in the context of determining, say, popular opinion (why do you think they can ask 1300 people out of the whole country and still predict who will win a national election, to within 3%?). But to **enumerate**, for apportionment purposes, one would have to make some very strong and precise assumptions about the density around which people were sampled. And how do you get that density? By doing a census ;)
So to get demographic breakdowns of (say) race, one could skip a total enumeration if the sampling was done well. But to determine (say) congressional seats, that would be another story.
One note on the math. Doubling in 10 years is not 10% annual growth. Percentages multiply, not add. 10 years of 10% growth gives about 2.59, or 159% growth. "Compounding". That is where the "7.2" rule comes from.
A fully concur with the point that some up front investment to increase participation may save money in the long haul. But I cannot vouch for any of the exact assumptions made.
Bob. You may know how to do math on the back of an envelope, but you don’t know finance. A 100% increase in 10 years is 7.2% annualized. Check it on your nearest financial calculator or go old school and look it up on an annuity table. Not outrageous when you consider the financial world of the 1990’s through mid 2000’s, talk was of 10% annual return for almost any investment.
This is a pet peeve of mine. Credit card companies and banks are able to fleece the general population because, as a group, we are financially uneducated. Present value, future value and compounding are all easy concepts that could be taught to high school students in a matter of hours (all based on nothing more difficult than fourth grade arithmetic).
So what you're saying Tired Boomer is that 100 divided by 10 = 7.2?
So the candy bar I bought in 1965 (don't laugh) for 5 cents that now costs me $1 really doesn't cost me 20 times what it did then.
//This is a pet peeve of mine. Credit card companies and banks are able to fleece the general population because, as a group, we are financially uneducated.
Maybe. But the first step is to get people interested in even doing the calculation. And I think it's nonetheless interesting that many of the people who are accepting the $80 figure for follow-up as a comparison to determine that $6 is a good deal, aren't that interested in finding out where the $80 figure comes from.
Maybe we need to learn the rule of compounding. But I vote for having more curiosity first.
//So the candy bar I bought in 1965 (don't laugh) for 5 cents that now costs me $1 really doesn't cost me 20 times what it did then.//
Ya, I bought ‘em for a nickel too (3 for a dime if they were stale).
You paid exactly 1/20 th FOURTY-FIVE YEARS AGO. The key is the ANNUALIZED percentage (apples to apples). Annualized percentages don’t make intuitive sense because of compounding (apples to oranges).
Don’t take my word for it. Ask one of the resources at your workplace, I’m sure a business or economics reporter will back me up.
Don't get me wrong. I understand what you're saying. If you start with $1 and the next year it's $1.10, it's gone up 10% that year. If it goes up to $1.20 the next year, it hasn't gone up another 10%, it's gone up around 9% (I didn't do the calculation on that). And if it goes up by 10 cents in a third year, it's gone up that year by 8+%.
So over the three years, it went up 30%, but it didn't go up 10% over three years.
But back to our story...
I haven't been able to get the Census Bureau to break down how they come up with the $75 per household figure of chasing down Census forms for people who didn't send them back.
But here's how I think they've determined that. I think they've determined that by the cost of chasing down the census forms using all the attempts they'd normally make. In 2000, I think this is involved something like three phone calls and three visits.
There's a flaw in that method. It assumes that the 7 million households who will send in their forms as a result of the letter, are as costly as the last 10 million households (out of 120 households) who have to be chased down for their forms.
But IF a single letter is enough to pry the forms from those 7 million households, isn't it unlikely that it would take the same effort to effect that as it would -- and will -- take to get the last of the households to send in THEIR forms?
Isn't it likely that 1 phone call would be enough... or 1 phone call and one visit would suffice? And, if so, doesn't that change the equation used to determine whether this is money well spent?
And then the question is how much does 1 phone call cost compared to the letter?
So far the Census has been goofy for me. I got the letter, then a week later I got the census. I filled it out and sent it in a week later. I got another letter telling me I'd received the census and to please fill it out. I ignored this thinking my census and this letter had crossed in the mail.
I got *another* census with an added letter telling me this was the second one they'd sent and please fill it in. I ignored this (probably shouldn't have). Anyway, Monday I got *another* letter telling me that they were sending me *another* census and if I didn't fill it out they'd contact me in person.
I will fill out the third census, but it is stupid, and frustrating.
Nice to hear you on the radio this morning. It got me thinking about how the Census is becoming a set-piece for political issues on various sides of the argument by the blog respondents. It seems the discussion like so many others has rapidly devolved into right vs left, as you pointed out.
Why is it that if a person doesn't implicitly trust the government to do something efficiently they are immediately labeled a right wing nut? Bob is right, we have a record national debt, and now a record deficit adding to it. The money isn't coming out of the couch cushions, and we can't break the future kids' piggy banks for it because we've left nothing in them. Asking where the money is coming from and where it goes does not make someone a Republican or "Conservative" or any other label. Wanting efficiency in government should be something EVERYONE is pushing for.
Conversely, wanting parity in health care or other social programs doesn't make someone a socialist. There are areas of service that the government can and should administer, without relying on the private sector. The "right" needs to accept that fact, and I would argue mostly they do aside from some very notable exceptions.
I've really started scratching my head on this one, because everyone on both sides of the political spectrum stands to gain from making the government more efficient. Instead, any time the right talks about efficiency or less government the left digs in and chastises for wanting to cut services and any time the left wants infrastructure or social service spending the right rants about waste and big government. We can have both if both sides just let go of some of the sacred cows.
I don't think it's always the "right" that questions spending. For example, yesterday DFLers questioned whether $86 million should be spent on the Moose Lake facility for sexually violent people and it was the Republicans who were pushing to spend more.
The problem comes when we race to compartmentalize people into a political definition on matters that may have -- and probably have -- nothing to do with politics at all.
The letters make as much sense as the ads on TV in January. Someone needs a reality check.
The advance letter is about saving money, believe it or not, and is standard practice in private industry. It saves money because people who receive an advanced letter are proven more likely to return the mailed survey. For every additional 1% of the population that returns the mailed form, rather than waiting to be visited perhaps multiple times by in-person census-takers, the government saved $85 million. So, if your cost calculations are correct, and if even one out of every hundred households return the mailed form who wouldn't have otherwise because of the advance letter, the government just realized a 100% return on investment. That's a pretty good deal.
Sometimes you have to spend money to save it.
I should add that the Census actually realizes an additional 6% of participation on the mail form when they send an advance letter, which saves $510 million. The advance letter pays for itself 11 times over.
It drives me nuts when people who don't know what they're talking about criticize government officials for "wasting money" when in fact they're making smart business choices. If you wonder why money is ever managed poorly in government - that's why. Officials who make good choices get lambasted by people who don't know any better.
//I should add that the Census actually realizes an additional 6% of participation on the mail form when they send an advance letter, which saves $510 million. The advance letter pays for itself 11 times over
At the risk of repeating myself, I'm rather surprised by how many people -- including journalists -- have accepted that figure without asking how it's calculated.
I've said to the Census Bureau, "prove it," and they haven't.
As I said above, the EASIEST people to get to return the survey, are people like me who were excited to think that's what was in the envelope.
The NEXT most-easy is the 7 million households that will send it in because they were sent a letter.
My theory -- and I'm left to guessing here because the Census Bureau won't answer the question -- that they come up with the $510 million figure by calculating how much time/overhead it takes to get the LEAST-likely respondents to respond. That could take three phone calls and three personal visits and that COULD cost $75 per household.
But that figure may not -- and logic says it DOES not -- apply to the targets of this letter.
And could we please get a little less lecturing about stupid we are for asking questions about how the money is spent. Because I'm noticing a lot of it is coming from people who are supplying more criticism than answers.
I posted on Census Bureau's FB page about the cost and they admitted to me that the actual cost was $88 million for the mailings.
You got to think like a government manager. Then it perfect sense. In the last 10 years the government has hired so many people for reasons other than performance who don't have a clue how to do their job and the only qualification was to be politically reliable. When you can't manage the big stuff what do you manage? The little stuff.
I pay a lot of taxes and for the most part I don’t mind paying. That is the price of being in our democracy. What I do mind is the money being wasted. I have worked for the Federal government for more than 30 years and I have seen it swing completely from one way of counting every penny to now where senior management couldn’t care less what they spend. When your measure of performance is how much money you spend there is no accountability.
Your price for how much that latter cost are way off. Bulk mail like this is not printed using any printer you would find in an office environment. It would have also not used rems of paper. It would have used Roll paper. Mailing, the most expensive part, would have cost less then a stamp per piece sent out.
Basically to send out 100 million letters would have cost a fraction of 35 million. If it was 7 figures I would be surprised as the printing industries prices are very low due to over capacity in the market at the moment.
Imagine if no one filled them out. And just sent them back.
As a sustaining member of MPR, I recently received a letter thanking me for my contribution of ...$0.00. I thought, "Woops! Why did they need to physically mail this anyway, I thought I was 'helping save administrative costs'." The next week I received a letter apologizing for the mistake of my $0.00 donation. I imagine I'm not the only one this happened to.
Same thing, smaller scale.
So thankful that I forgot my eyeglasses this evening. With blurred vision there's still a chance that I might be "seeing" more zeros than I thought. I'd heard the total cost to be $86 million; this includes effort to visit homes for those who had not completed forms. THAT is ridiculous! The census envelope was quite obvious to me -- if needed, they could have painted it red or purple to help it stand out for crying out loud. WASTE.
This is why the gov'mnt is saying that social security won't be around in short time..Iam willing to bet all this money is coming from ss, but they won't tell that.
I recived census form twice and 3 other letters. I'm so up set how the government waste tax payer money like that.
Cost of living increase!!!: I pay more for medical insurance, more for my apartment expenses, transportation..............On top of that my salary got cut. It's getting really hard!
During this hard time goverment is wasting millions dollars on seding extra notices and froms. Is that how they create job?
People are sitting in the White House don't feel our pain for sure!
We got a letter saying we were going to get a Census form. Then we got another letter saying we should fill out the census form we never got. Then we got a postcard saying how important it was to fill it out. We finally called and requested a form. We filled it out. Then a NRFU enumerator was assigned to fill another one out. Many of our neighbors did not get Census forms or letters.