Mintage question.

I've been thinking about this for a while...
How accurate are mintages? I understand that the mintages of early US coins are based on the amount of coins that were delivered from the mint, and sometimes the delivery of coins minted one year will end up in the deliveries of the following year. Therefore, the actual mintages are really anyone's guess. For example, no one really knows how many 1804 Draped Bust cents were actually struck. And the concensus of that number has changed over the years. My 1962 Redbook states that 756,838 were struck, but my 2010 edition estimates 96,500. An estimate that comes from the best guess of survival rates. But what about modern coins? Are the mintages based on the same principle of delivery? What about multiple struck coins? Or bonded planchets? Capped Dies? Does the Mint count how many times all of the dies struck something (i.e. anything?), and that's the mintage? Are blank planchets that leave the mint counted in the mintage, even though they weren't struck? The Mint claims that exact numbers of coins are struck—like 4,290,000,000 2006 1c pieces, or even 6,426,650,571 1993-D 1c pieces—but how accurate is that? Let's discuss.
How accurate are mintages? I understand that the mintages of early US coins are based on the amount of coins that were delivered from the mint, and sometimes the delivery of coins minted one year will end up in the deliveries of the following year. Therefore, the actual mintages are really anyone's guess. For example, no one really knows how many 1804 Draped Bust cents were actually struck. And the concensus of that number has changed over the years. My 1962 Redbook states that 756,838 were struck, but my 2010 edition estimates 96,500. An estimate that comes from the best guess of survival rates. But what about modern coins? Are the mintages based on the same principle of delivery? What about multiple struck coins? Or bonded planchets? Capped Dies? Does the Mint count how many times all of the dies struck something (i.e. anything?), and that's the mintage? Are blank planchets that leave the mint counted in the mintage, even though they weren't struck? The Mint claims that exact numbers of coins are struck—like 4,290,000,000 2006 1c pieces, or even 6,426,650,571 1993-D 1c pieces—but how accurate is that? Let's discuss.



0
Comments
The deliveries were of good coins, so they excluded coins that didn't meet the standards (errors, etc) and were never delivered. Also, in at least one case in 19th Century New Orleans, 25,000 half eagles were later condemned and melted (after delivery) and the coins were not included in mintage figures.
The Philadelphia and branch mints provided monthly production statistics, which were used to calculate the mintage numbers that were published in the Mint Annual Reports.
Early in the Mint's history, the Mint staff paid more attention to the number of coins produced rather than what date was on the coin, which leads to the confusion you mentioned for the 1804 cents (also, die steel was scarce, so dies were sometimes re-used; this practice was abandoned later in the 19th century). Mint historians, such as RW Julian, have compared die delivery statistics to coin production statistics which has cleared up some of the confusion.
Late in the 19th century, the Mint Annual Reports started publishing comprehensive statistics of coin production at each mint. In one of these reports (I forget which one off the top of my head) the introduction to the statistics details the trouble the Mint staff made to make the tables reflect the date on each coin, rather than just the year of production.
If you wish further information, Roger Burdette (member ATS) can discuss the monthly production records that he's seen and you can find some of RW Julian's articles that provide detailed information on mintages (he is - or was- a member here, so he may comment himself). Also 19th and early 20th century Mint Annual Reports are available on Google Books and the later 20th century reports are also available on the Internet - I can give you an address, if you want it.
I don't know much about the post-WWI Mint, but I expect that modern coin mintages are based on deliveries, too.
edited to add: some clarifications.
Check out the Southern Gold Society
when you're producing billions of a coin, what do you count?
is there a computer controlled and read counter on the die press that counts strikes?
do they subtract for multiply struck coins?
or maybe do they count planchets in?
do they subtract for condemned coins (coins they discard) ?
etc.
questions to the infinity of the imagination.
beginning of the year being included with the earlier year rather than the date on the
coin. 1971 quarters have a low reported mintage of about 110 million but there never
seem to have been so many in circulation. In some cases one can speculate that coins
were struck in Denver without the mintmark.
I suspect that clad mintages tend to be very close with the exception of the '71 quarter.
I do think of mintages though as "calculated" rather than counted; they know to the
pound how many coins were palletized in a given year.
<< <i>the question goes deeper than that.
when you're producing billions of a coin, what do you count?
is there a computer controlled and read counter on the die press that counts strikes?
do they subtract for multiply struck coins?
or maybe do they count planchets in?
do they subtract for condemned coins (coins they discard) ?
etc.
questions to the infinity of the imagination. >>
Exactly. How much of the 624,500,000 mintage does this hold: 3, 1, or 0?
<< <i>
Exactly. How much of the 624,500,000 mintage does this hold: 3, 1, or 0?
>>
LMAO