Understanding how program funding works for JIT Gateway

[pb-cons]

I need to confirm some assumptions about JIT Gateway funding.  I've gone through the relevant guides but I'll need some help to make it stick.

Assume that I setup a Marqeta integration using a JIT Gateway approval webhook.  I believe that funding those purchases goes something like this:

• I'm issuing cards based on a Card Product. That Card Product has a unique Program Funding Source associated with it.
• That Program Funding Source is connected to an account external to Marqeta.
• When a transaction is approved by the gateway, Marqeta will draw funds in real-time from the external account tied to the Program Funding Source that exactly match the authorized amount at the time of the approved authorization. These funds are added to the GPA for the cardholder.
• If a transaction is cleared for a larger amount, Marqeta will draw funds in real-time from the external account tied to the Program Funding Source that exactly match the additional amount at the time of the clearing.  These funds are added to the GPA for the cardholder.

Questions:

1. Is my description above accurate?
2. Can a Program Funding Source be used by multiple Card Products?
3. If a Program Funding Source can be used by multiple Card Products, what are the drawbacks to this?  Is reconciliation more difficult for instance? Would a 1-1 between a Card Product and a Program Funding Source be considered best practice?
4. In a JIT Gateway scenario, is the Program Funding Source the only way that funds can enter a GPA (other than a refund/reversal with refunds_destination=GPA)?
5. What happens when a transaction is cleared for a lower amount than the authorization, either at clearing, or as the result of an Advice transaction?  Is this also controlled by the value of refunds_destination?
6. What happens should Marqeta be unable to draw funds from the external account associated with the Program Funding Source?  Can I simulate this in the sandbox? 

[jcook · Accepted solution]

Hi @pb-cons! See my answers below:

1. Yes this is correct.
2. Yes, a Program Funding Source can be tied to multiple Card Products.
3. The drawbacks would be if you need to parse the data or keep it separate from the data of the other card products, if it can’t be parsed on your end. Or if the card products need to draw funds from separate sources. 
4. Yes, correct. 
5. It is not controlled by “refunds_destination”. If this happens and “enable_partial_auth_approval” is set to “false”, then the transaction will decline. If it’s set to “true” then the card will clear for a lower amount. Typically in this scenario, you are expecting multiple clearings that add up to the original amount of the authorization. For example, when you buy goods from Amazon, the authorization comes through for all of the goods, but you get charged as the items ship. If one of those items was canceled, then it would appear as a reversal. For Advice, you would see a message reflecting “amount to be released” in the clearing that directs you to release those funds from the cardholders account. 
6. You would get a decline reflecting “insufficient funds for program reserve account” and unfortunately we do not have the ability to simulate this at this point in time.

[pb-cons]

Great, thanks again @jcook. I'll probably have some follow-up questions on these points.

[Rajgupta]

In above scenario, Is program funding source a checking account (FDIC insured) or more like a card account (non FDIC insured and not a checking/savings account) that can only be used for the credit cards spend?