. >> we didn't explicitly use the e ternl cost data. we used the data we gathered through our survey. having said that, we did look at various external cost studies associated with issuer costs. there are various problems, including the cost accounting systems can differ allot across firms. and in addition, whether or not a firm outsources things to a third party may affect their ability to identify certain types of cost components. and finally we recognized as we were performing the incentive effects associated with reporting for respondants enganged in the exercise we were looking at. but when we did look at the external studies, performed by consultants and processors and payment card networks, the numbers they got were largely comparable to what we received from our su va. there are caveats associated with that in they didn't survey the same issuers we surveyed. the level of granularity, we don't know how well that compares. at least it's sort of an aggregate level. their estimates are largely comparable to what we got. which makes u

. >> we didn't explicitly use the e ternl cost data. having said that, we did look at various external cost studies associated with issuer costs. there are various problems, including the cost accounting systems can differ allot across firms. and in addition, whether or not a firm outsources things to a third party may affect their ability to identify certain types of cost components. and finally we recognized as we were performing the incentive effects associated with reporting for respondants enganged in the exercise we were looking at. but when we did look at the external studies, performed by consultants and processors and payment card networks, the numbers they got were largely comparable to what we received from our su va. there are caveats associated with that in they didn't survey the same issuers we surveyed. the level of granularity, we don't know how well that compares. at least it's sort of an aggregate level. their estimates are largely comparable to what we got. which makes us think our numbers are not radically off or the