Hi, Evan,
 
I may not disagree with your arguments.  However, regarding calculating the "costs", that is only next to impossible.
 
Calculating the "costs" of the new gTLD program, is very similar to the telecom industry adding new services.  Exactly what would be the costs for a particular new service?  This could range anywhere from ZERO to infinity.  The reason for this is, without any exception, any new services would always use the existing network, which should be the major "cost" of this new service.  However, regardless of there is such a new service or not, the network is already built for the "pailn-old-telephone-service" (POTS). All the costs for building it is already spent, and there is no way to retrieve it.  Thus, it is known as "sunk costs", and is already paid for by the tariff for POTS.
 
Therefore, what is the real cost for any additional new service?  In theory, it should only include any additional equipment attached to the network specifically for this new service but for nothing else.  However, there are still problems with calculating the per unit cost of the new service.  For example, how many people are going to use this new service?  Also, for how long?  This is pretty much like a digital camara priced at $100.  If you use it to take exactly 100 photos, the cost of each photo will be $1.  If you take 1,000 photos, each will cost only $0.1.  However, if you took only one photo and it is broken, this single photo will cost $100, while if it was stolen even before you could use it to take pictures, the cost to you for each photo with this camara will be infinite.
 
From the above exmples, I believe you can already see the complexity of calculating the "real cost" of the new gTLD program.  Simply said, I do not believe in any claims of the "real cost", becuase it does not exist at all.
 
Now back to our original issue: What to do with the excess funds collected via auctions?  If the original intention was nothing but to recoup the "costs" of the new gTLD program, and if this is how the USD $175K was calculated (or guessed) to begin with, then this "excess" funds is fully justified to be beyond ICANN's intention of recouping the "costs".  Therefore, providing a refund to the payees of thier $175K would also be fully justified.
 
Thank you again.  Further discussions are most welcomed.
 
Kaili
 
 
----- Original Message -----
From: ja.bhenda@gmail.com
To: Evan Leibovitch ; Kan Kaili
Cc: ICANN At-Large list
Sent: Saturday, May 13, 2017 4:06 PM
Subject: Re: [At-Large] Auction Proceeds - where we are and what you can help

Thanks Evan

Your arguments clarifies the matter

Jean Benda Nkurunziza
AtLarge-ISOC Rwanda President
+250731000006

Sent from Bhenda Latest Tech Robot


-------- Original Message --------
Subject: Re: [At-Large] Auction Proceeds - where we are and what you can help
From: Evan Leibovitch
To: Kan Kaili
CC: ICANN At-Large list


On 13 May 2017 at 01:05, Kan Kaili <kankaili@gmail.com> wrote:

Besides all the suggestions for fund usage, I would suggest another way to "use" the money.  That is, to REFUND all the applicants who paid for applying new gTLDs.

​The original price was set by policy of cost-recovery, at the price that it was because there was no precedent and that ICANN needed to build-in contingencies for legal challenges and recoup previously-spent expenses to design the​ gTLD program.

​ICANN has never calculated the actual cost of delivering the program, and until it does the amount of a refund can not be known. I think that the result of such research would be surprising, that the gap between actual cost and anticipated cost is far less than might be dreamed by the domain industry.

Remember that the issue under debate is not the disposal of excess funds from the gTLD application fees over real costs, but of auction proceeds​ gained well in excess of those fees by applicants willing to pay an even higher premium to get specific strings.

 If I remember correctly, ICANN collected USD $175K per application, and recognized as a hefty price to pay even by ICANN itself.  This could be a threshold that prevented some or many potential applications.

​Perhaps a lower fee for future rounds based on the experience of previous wounds would address this situation, but a retroactive refund would not.

​- Evan​