col-xs-12
col-sm-12
col-md-12
col-lg-12
col-xs-12
col-sm-12
col-md-12
col-lg-12
videoType
dynamic-media
videoDmUrl
https://delivery-p136806-e1377785.adobeaemcloud.com/adobe/assets/urn:aaid:aem:25981677-e362-42ad-a250-d9172397095b/play?assetname=Overlay-Services-Webinar-Recording-20260128-080546.mp4
keyFrameImage
videoDescription
Fast payments are characterized by the instant transmission of the payment message and by the immediate availability of funds to the beneficiary on a 24/7/365 basis. But authorizing, clearing, and settling a payment are but one layer of a robust fast payment system. Atop them lies a variety of overlay services that can improve product value, prevent fraud, broaden adoption, and drive the overall success of the fast payment system.
timestamp

00:04 Hello everyone and welcome to the World Bank's webinar series on fast payments.

00:08 My name is Colin Coulter,

00:10 and I'm a consultant with the World Bank based in Washington DC

00:14 Project Fast,

00:15 or frictionless Affordable Safe Timely Transactions,

00:18 is a flagship project of the World Bank sponsored by the Gates Foundation

00:22 and which aims to accelerate the adoption of

00:24 fast payments in low and middle income countries.

00:26 Now just a reminder that you can find all the resources that

00:29 the World Bank has been producing on the fast payment work,

00:32 including the recordings of these webinars under Project Fast.

00:36 On a dedicated website,

00:38 fastpayments.worldbank.org,

00:41 today we are continuing with our series this time

00:43 focusing on prominent overlay services and fast-paced payment systems,

00:48 and we really have the perfect guest to lead us through that conversation,

00:51 Claudio Cerrazani.

00:53 Claudio's career spans 25 plus years of extensive leadership,

00:57 management,

00:58 and global business experience in payment systems.

01:01 He is a subject matter expert

01:03 on real-time payments,

01:04 high value in retail payments processing,

01:06 central bank digital currencies,

01:08 and the clearing and settlement of securities and equities.

01:11 To date,

01:11 Cloudy has worked with more than 60 countries' payment system infrastructures,

01:16 providing business and technical advice.

01:18 He has also personally shaped the transformation

01:21 of more than 30 national payment system implementations.

01:24 In 2018,

01:25 Claudio launched DGT Solutions

01:28 designed to empower central banks,

01:29 financial institutions,

01:31 and FinTechs to undertake complex payment system transformation initiatives.

01:36 And since then,

01:37 the DGT Solutions team has supported payment

01:39 systems development in more than 20 countries.

01:42 Such as the UK,

01:43 UAE,

01:43 Saudi Arabia,

01:44 Montenegro,

01:45 Albania,

01:46 Kosovo,

01:47 North Macedonia,

01:48 Maldives,

01:48 Lesotho,

01:49 Tajikistan,

01:50 Gambia,

01:51 Vietnam,

01:52 Palestine,

01:52 etc.

01:53 The list goes on.

01:54 So we really have the perfect guest today to talk us through

01:58 prominent overlay services and fast payment systems.

02:01 Claudio,

02:01 thank you so much for joining,

02:03 and the floor is yours.

02:06 Thank you,

02:06 Colin.

02:07 Thank you everyone.

02:09 Today we will focus on overlay services,

02:14 so

02:16 overlay services

02:18 that actually are around what is the fast payment system.

02:24 What are overlay services?

02:26 They are,

02:27 of course,

02:27 value added features that are,

02:30 let's say,

02:31 layer on the FPS core infrastructure are

02:35 completing,

02:36 let's say the FPS core infrastructure.

02:38 We will say we will see how

02:41 and

02:42 what actually,

02:43 what is the scope of the various services is basically to

02:47 make an instant credit transfer.

02:50 And

02:51 cover use cases

02:53 that a normal push payment could not cover.

02:56 A normal push payment with

02:58 an account identification could not cover.

03:01 So

03:02 that actually is allowing

03:05 an FPS

03:06 to

03:07 expand its reach and

03:10 cover multiple use cases.

03:12 And of course what the benefit is that

03:15 of course it's promoting adoption

03:18 and

03:19 competition,

03:21 innovation because it actually creates

03:24 a layer of

03:26 business

03:27 functionality that can

03:29 be used to create a new service and

03:32 of course as a consequence of inclusion because it can reach.

03:35 Any kind of

03:38 a person,

03:39 any kind of business.

03:43 These are just a list of common overlay services.

03:46 Actually,

03:47 the list is increasing

03:49 and you can see

03:51 a request to pay,

03:52 confirmation of payee,

03:54 proxy directory

03:56 using a flies,

03:58 biller directory

04:00 that actually

04:02 integrating billing system with the fast payment system,

04:06 tokenization.

04:08 Let's say

04:10 changing the account number in random tokens in order to

04:15 for privacy

04:17 white label application

04:19 enabling the PSPs to use

04:23 a pre

04:26 adapted preformed white label application

04:29 for management,

04:30 the possibility to

04:32 For the FPS operator to

04:36 install a centralized management system and share

04:40 the service with the PSPs,

04:43 core code Man,

04:43 a centralized system that can issue

04:46 core codes.

04:47 Also the FPS,

04:49 I

04:49 think as we discussed in other webinars,

04:51 can also be

04:52 leveraged to

04:55 do an open banking app

04:57 now today.

04:59 We will just focus on the,

05:01 let's say,

05:01 what is the prominent of services

05:04 that are requested to pay

05:05 a confirmation of pay.

05:10 There are general capabilities that are common

05:14 to these two overlay services request to pay and confirmation of the

05:19 Messaging standards.

05:21 Currently ISO 2022,

05:23 it's dominant.

05:25 It's of course it's de facto standard.

05:29 Some of jurisdictions are still not using

05:32 ISO,

05:32 but it's more an historical reason.

05:36 And

05:38 let's say all the most recent

05:40 implementation and of course the future implementation

05:43 will adopt ISO 20022,

05:45 very flexible

05:46 and with specific messages

05:49 for requests to pay and confirmation of pay.

05:52 Operational hours,

05:53 these are the service of a fast payment system

05:56 and of course they must follow

05:59 the same

06:00 availability

06:02 standards,

06:03 so they must run 24/7.

06:06 API

06:07 more and more overlay services are based on API more than messaging.

06:16 So and we will see that

06:18 let's say that the trend in the market is that specifically for the services.

06:24 And

06:25 the use of API is becoming

06:29 predominant.

06:32 The use capability,

06:33 the possibility to use these

06:37 services using aliases,

06:39 so both a request to pay and confirmation of pay as the possibility to use

06:45 aliases instead

06:47 account identifications.

06:49 Access channels.

06:51 These services must be,

06:55 must be

06:56 available on all channels,

06:58 same as

07:00 fast payment system participation.

07:02 Now participation,

07:03 it's another.

07:06 Let's say specific um.

07:09 Uh,

07:10 of,

07:10 uh,

07:10 the,

07:11 the confirmation of PNR requests to pay that actually can have uh

07:15 It can be treated as a separate service to the first payment,

07:19 although they are complementing

07:22 the fast payment system and they are enabling fast payment system

07:27 use cases,

07:28 they can also be seen as a separate service,

07:30 so the participation

07:32 to this overlay service can also be

07:35 separated.

07:36 Let's say that participant to overlay service can be

07:41 more

07:42 or less than

07:43 the participant of the FPS.

07:46 Business continuity,

07:47 of course,

07:48 exactly this is because it's coupled with

07:52 a fast payment transaction.

07:54 You must

07:55 take care of the business continuity

07:57 with the same care

07:59 as the fast payment system.

08:05 Now,

08:05 first,

08:06 let's start with request to pay.

08:08 What is the goal of request to pay?

08:11 Streamline payments,

08:12 enhance flexibility,

08:13 improve cash flow,

08:14 and expand FPS adoption,

08:16 um.

08:18 Motivation are many and

08:20 let's say you can see it here in the slides,

08:23 but what I want to also to stress that

08:26 the request to pay

08:27 is

08:29 A really powerful tool.

08:32 Request to pay protocol.

08:34 It's

08:35 extremely flexible,

08:37 so it

08:39 can really be used

08:41 also beyond the use case that I will now show

08:45 because actually it's open to creativity from the PSP

08:50 also to create

08:51 new services.

08:54 These are let's say basic request to pay transaction flow.

08:59 There is of course

09:01 an engagement with the payer and the payee on

09:04 buying

09:05 goods,

09:05 services,

09:07 and point number 1

09:10 when agreed this exchange,

09:12 the payee

09:13 can send

09:15 a request to pay to the payer.

09:17 It's

09:18 sometimes also called the pool.

09:20 Payment.

09:21 So

09:22 the payee is pulling

09:24 the payment.

09:25 The payer

09:26 can offer accept,

09:28 reject,

09:29 and then we will see there are also other menu options

09:33 depending on the use case.

09:34 And let's say if you accept

09:37 now the

09:39 step 4,

09:40 there is the credit transfer that starts as a normal,

09:44 let's say credit transfer,

09:45 instant credit transfer.

09:49 Uh,

09:50 these are some capabilities that

09:55 should be taken into consideration where implementing a request to pay

10:02 now

10:03 uh

10:03 just in a

10:05 Let's say in terms of experience of implementation,

10:10 um,

10:12 not

10:13 the request to pay a different maturity and then of course

10:17 not to request to pay.

10:19 As all the capabilities that we can see here.

10:22 This is more

10:24 a suggestion for a way forward.

10:26 These are

10:27 also including sophisticated capabilities that are

10:30 enabling

10:32 a more sophisticated use case on request to pay.

10:35 Also,

10:35 request to pay can be very simple.

10:37 Anyway,

10:37 I just mentioned,

10:38 of course,

10:39 the messaging

10:40 as it was pretty much standard.

10:42 Pane 1314 is the ASU standard.

10:47 One thing that it's important in SLA management,

10:50 so the capability

10:51 to

10:53 be sure that the end to end transaction,

10:55 both the request to pay and the linked payment

10:58 are enforced,

11:01 are enforced by the system.

11:04 The idea is that the request to pay can support multiple instruments.

11:08 Another thing that it's we are speaking about first payment,

11:11 but the request to pay as a tool can be also seen as a separate service.

11:14 So

11:15 a request to pay

11:16 can also

11:18 trigger a payment that actually is not a fast payment.

11:23 So in some implementation,

11:24 the request to pay,

11:25 it's actually

11:26 A wider than the first payment itself.

11:30 So after request to pay,

11:32 it's not mandatory to send an instant payment,

11:35 but you can also

11:37 complete with different

11:39 instruments.

11:40 So

11:41 it can be connected or not.

11:44 Let's say for the

11:46 maximum flexibility it's better to keep it open.

11:50 And

11:52 the possibility to monitor the processing time in the different steps and legs

11:56 and

11:57 of course the possibility to use the same infrastructure

12:02 to handle

12:03 both the payment trail and the new financial message as also the.

12:09 A request to pay and confirmation of the app code,

12:13 so payment,

12:14 nonpayment messages,

12:15 sometimes financial,

12:15 nonfinancial messages.

12:18 Fraud,

12:18 we need to also

12:20 understand that the request to pay

12:22 needs also fraud prevention because it's a pool payment

12:26 and being a pool payment it's also open to let's say the same type

12:31 of fraud that

12:33 an instant credit transfer

12:35 may have.

12:36 So

12:37 important focus on

12:40 fraud prevention

12:42 and there are multiple ways.

12:45 Transaction limits,

12:46 and I think it's also to define

12:49 the possibility to define.

12:52 Also rules that are that are related to transaction limits,

12:57 how a request to pay can be used in different use cases,

13:01 what are,

13:02 what are the limits you can put depending on where the request to pay is used.

13:08 Retailer measure is different from a request to pay using business to business

13:13 or in a billing

13:15 situation.

13:16 Of course,

13:17 another thing that it's important is a granular message status manager,

13:21 so the possibility to

13:24 categorize the request to pay in pending paid,

13:27 not paid,

13:27 rejected to be able to actually

13:32 track the life cycle of requests to pay also

13:36 connected

13:37 to the related payments

13:40 there can be one or multiple.

13:46 And

13:47 these are some of the business capabilities that are important,

13:51 so the possibility to have a single request to pay,

13:53 a recurring request to pay,

13:55 also a bulk request to pay.

13:57 You can see that in

13:58 some government

14:00 or bill application,

14:02 having the possibility to have a bulk

14:04 request to pay can be important.

14:07 The possibility to have a flexible acceptance option.

14:10 So for the payer accept now or accept later.

14:16 Possi to have the flexible payment mode.

14:20 Pay now.

14:20 I can pay later.

14:22 I can make partial payments,

14:24 pay some of what was requested.

14:28 And so these are the business capabilities that are very,

14:32 very interesting if you implement a request to pay.

14:34 But if you get a cancellation,

14:35 the possibility also to cancel

14:38 a request to pay

14:39 in certain use cases when your

14:42 request to pay is used to agree,

14:44 let's say when a request to pay is used for a similar direct debit.

14:49 You need also to be able to cancel

14:53 this

14:54 agreement that was

14:56 taken with a request to pay,

14:58 and of course you can also use the request to pay for other

15:02 specific.

15:05 Meaning

15:06 you can have multiple requests to pay to agree between payer and payee.

15:10 You can use the request to pay to send reminders

15:14 of payment

15:16 also

15:17 to trigger

15:19 OTP or approval to give extra security.

15:25 I'm speaking about this,

15:26 but there are actually,

15:27 it's not actually theory.

15:30 They put in practice all these business capabilities.

15:33 Make the RTP much

15:35 very,

15:36 very powerful.

15:39 These are

15:41 some use cases as I told the use case that the request to pay can cover are many

15:48 here I'm just,

15:51 let's say

15:52 listing the basic ones

15:54 that is one of FTP

15:56 that it's actually

15:59 the pay and FTP you accept reject,

16:02 and

16:03 basically it's giving

16:06 Full

16:07 control to the payer,

16:08 uh,

16:09 very,

16:09 very transparent,

16:11 uh,

16:12 and

16:13 can be even real-time,

16:15 um.

16:17 It's,

16:18 it's an idea for adopt payments

16:21 or

16:22 govern to

16:24 payment govern to personal outreach,

16:27 and

16:28 it's actually

16:29 one of the basic use cases that

16:32 let's say that an RTP can cover.

16:35 Then we have the recurrent RTP that actually

16:38 probably is the most interesting use of it.

16:40 Utilities,

16:41 rent,

16:42 bills,

16:42 telecom billing subscription,

16:46 membership insurance,

16:47 it's very powerful.

16:50 It can of course

16:52 actually be a valid alternative to the direct debit.

16:56 Actually it's much more flexible than a direct debit,

16:59 and

17:00 it is also

17:03 can be also more transparent

17:05 for the for the payer.

17:08 So it can also enhance.

17:10 The way direct debit is done.

17:12 And create a lot of flavors,

17:15 let's say,

17:16 of

17:17 let's say direct debits.

17:19 Then there is the last that I want to mention point of interaction request to pay.

17:24 Why it's important,

17:25 it's because

17:27 a request to pay if used

17:29 on a POS

17:31 can

17:33 mimic

17:34 a typical card transaction.

17:37 Usually when I mean when you want to go on the merchant,

17:41 the most common way it's

17:43 the payer,

17:44 the merchant started the transaction,

17:46 so the payer exchanged,

17:48 insert the card,

17:49 exchange some data with NFC.

17:52 And the merchant starts the transaction.

17:56 Of course in a in a fast payment system that is a push payment,

18:02 you typically

18:04 the basic use case on the merchant side

18:08 is

18:08 actually the opposite.

18:10 The merchant show a core code,

18:11 I read the core code,

18:13 and I push a transaction.

18:15 Using the request to pay,

18:17 in reality,

18:18 unlock

18:20 more use case that can be done

18:23 with an instant payment,

18:24 instant payment system,

18:26 an instant credit transfer,

18:27 because

18:28 with the request to pay,

18:30 for example,

18:31 let's say that I have my,

18:33 now I'm making it simple,

18:34 but actually it's much more complex,

18:36 but

18:37 if I have a card with my account ID and I

18:41 tap on the POS.

18:44 the POS,

18:45 actually the merchant

18:47 start a request to pay.

18:49 That goes to my bank.

18:52 The bank

18:53 basically can

18:55 there are some,

18:56 let's say,

18:57 rules.

18:58 Consider that a pre-approved transaction

19:01 if you have the money in the account,

19:03 goes to the confirmation and the merchant,

19:06 of course,

19:07 gives

19:08 you the goods.

19:08 So that's actually

19:10 now speak just one of the cases,

19:12 but give the idea

19:14 on what request to pay can unlock

19:17 when used on a

19:20 on a point of interaction.

19:22 Uh,

19:23 Claudio,

19:23 this is,

19:24 I really appreciate the breakdown on the kind of specific use cases.

19:27 I don't want to slow us down,

19:27 but I'm curious because in earlier slides,

19:30 you're talking about the importance of how this

19:32 can help build trust in the payment system,

19:34 how the importance of like fraud protection stuff,

19:38 um,

19:38 of course,

19:39 wrapped up within

19:40 an overlay services program.

19:42 I'm curious,

19:43 back on that slide,

19:44 are any of these

19:45 use cases.

19:47 Do you think they are

19:49 maybe more promising in building trust when

19:51 people are first launching one of these systems

19:54 than any of the others?

19:55 I'm just thinking about

19:56 it,

19:57 it's fully open.

19:58 Everyone can kind of use this.

20:00 If you're building an FPS launching or launching an RTP system,

20:04 which of these is kind of the strongest way to build trust in users?

20:11 OK,

20:13 every,

20:13 every kind of services

20:16 must be

20:17 introduced with care

20:19 because

20:20 it gives you the potential to have new use case,

20:23 but it's also opens up

20:25 to new

20:26 threats.

20:28 And

20:29 request to pay you can

20:31 may understand if you,

20:33 for example,

20:33 if you allow everyone

20:35 to send a request to pay to everyone else

20:40 and you may be receiving

20:43 a request to pay

20:44 from you don't know

20:47 and

20:48 you need to

20:50 before introducing a request to pay,

20:52 you need to analyze the specific use case and put

20:56 Very specific rules

20:58 on it

20:59 in order to create the trust.

21:01 If I would like to,

21:02 let's say start creating trust,

21:04 I would

21:05 start with,

21:06 let's say more utilities,

21:08 more like

21:10 business to person,

21:12 where in reality,

21:13 of course it's more controlled because

21:17 you can control who is generating the request to pay.

21:20 And a lot and you actually give more

21:25 transparency and more data to the payer.

21:29 And

21:30 to check what he has to pay,

21:34 so I send you an invoice.

21:36 I send you a subscription.

21:37 I can see,

21:38 I can check

21:40 it's good that it's what I was expecting,

21:42 and I can pay immediately

21:44 actually in the same in the same transaction or later.

21:48 I mean it depends probably

21:51 a utilities bill,

21:52 it would be have a pay later

21:55 functionality,

21:55 you know,

21:56 I want to check and

21:57 I can pay.

21:58 And I can probably

22:00 sometime put a time.

22:01 I mean,

22:02 OK,

22:02 I agree I pay at this date.

22:05 The sophistication of request to pay and the flexibility,

22:08 it really

22:09 can be very,

22:10 very high and give you all these,

22:11 and I think that it's what a customer

22:14 will like most,

22:16 and you are in a safe situation,

22:19 then little by little

22:20 can expand on other use cases.

22:24 Thank

22:25 you.

22:25 That's very helpful.

22:25 It's almost like the,

22:26 the real value creation is that transparency

22:29 that then builds the trust,

22:31 choice.

22:33 Uh,

22:33 thank you.

22:33 I don't want to slow us down.

22:35 OK,

22:36 now some legal regulatory economic aspect

22:39 aspects on the legal and regulatory,

22:43 of course it's important to

22:46 specify when a request to pay is a binding obligation,

22:50 meaning when,

22:51 and that is of course it's done with

22:55 and enforced with

22:58 request to pay scheme rules

23:01 so very clear.

23:02 Um,

23:04 very,

23:05 very,

23:06 very clear definition of the obligation

23:09 of the participants in this overlay service.

23:14 Then of course also

23:17 Ensuring customer potential,

23:19 so a very clear dispute resolution mechanism,

23:22 very clear liabilities.

23:25 between the participant who is paying for if something goes wrong

23:29 and then and that this can also be

23:32 different

23:34 depending on the use case that you are implementing.

23:37 If the request to pay is implemented in a point of interaction,

23:40 you can have

23:42 different obligation,

23:43 different dispute resolution mechanisms than if you are

23:46 doing,

23:48 let's say,

23:48 a request to pay

23:50 for a utility bill.

23:53 In reality,

23:54 every use case

23:57 most probably would have

23:59 its own rules.

24:01 In request to pay,

24:03 you don't actually have

24:05 a request to pay

24:06 as a protocol and scheme rules of request to pay.

24:10 Actually,

24:10 request to pay,

24:11 it's very different

24:14 in the way it's used.

24:16 I mean if I use a use case

24:18 it can be a totally different

24:21 business process and obligation than in another use case that must be very,

24:25 very clear.

24:27 So

24:28 it can be integrated

24:30 in the

24:31 fast payment system rule.

24:33 It's usually suggested to be separate exactly for the reason I was just saying

24:38 because the

24:39 the RTP rule can be actually quite complicated

24:43 and quite different depending on use case to use case.

24:46 In terms of fee models,

24:47 actually

24:48 there are many,

24:49 many models,

24:51 but usually because the request to pay is triggering

24:55 a fast payment system,

24:57 a fast payment transaction,

24:58 sorry.

24:59 And

25:00 you usually,

25:02 let's say the approach is to make it

25:05 free or very low

25:08 in order to create

25:11 more adoption because actually it's

25:14 pulling more transaction

25:15 on an instant payment system

25:17 and again depending on the use case

25:20 on a P2P most probably

25:22 you should go for no no fee.

25:25 Of course in a business situation you may also let

25:30 the participant,

25:32 let's say be more free or

25:34 charge

25:35 a higher fee.

25:36 So that's also

25:38 another thing.

25:38 The fee model can vary.

25:40 On a specific use case where they request to pay is used.

25:46 Let's see the timing.

25:47 OK,

25:47 I don't want to

25:49 too much on the request to pay benefit the challenges,

25:52 uh,

25:52 I believe,

25:53 uh,

25:55 every

25:56 request to pay has a lot of benefits that I can,

25:59 I can

26:00 really highlight that.

26:02 But of course,

26:04 again,

26:05 this new service must be introduced with care

26:09 because depending on the use case,

26:10 depending on is implemented.

26:13 You may have issues of privacy.

26:16 You may issue of fraud.

26:19 Sometimes you have,

26:20 you can also confuse

26:23 the payer.

26:26 So the other thing for for business for PSP

26:32 can be complex

26:34 depending on how the protocol is created.

26:36 So you may have additional cost.

26:40 Liability can be different.

26:42 So also the business process that you need to put in place can be complicated.

26:48 You need a specific,

26:51 let's say,

26:52 a regulation or rules around it

26:55 that

26:56 must be followed.

26:59 And you need to

27:01 have clearer legal liability rules that again

27:05 when you think of instant payment system,

27:08 it's one rule when clarify that and the request to pay can be

27:12 actually multiple

27:14 depending on

27:15 how you use it.

27:20 I

27:21 mean here we just

27:23 put some kind of example of request to pay.

27:26 In reality,

27:27 despite

27:28 the

27:29 power of request to pay,

27:33 the user request to pay,

27:34 it's not so wide as it

27:37 should be in my opinion,

27:40 and I think it's because of

27:45 Let's say

27:46 some sometimes being having something very powerful makes it difficult to

27:52 let's say apply it to a specific case and

27:55 break down all the business process,

27:57 understand the

27:58 make the

28:00 consumer understand the value and the businesses.

28:04 Anyway,

28:05 let's say that the request to pay,

28:06 if you can see,

28:07 was launched in the most prominent

28:10 region and jurisdictions.

28:12 So it's something that it's

28:15 in my opinion,

28:15 a key for a fast payment system to evolve.

28:23 These are some of the cross-cutting

28:25 success factors.

28:27 Regulatory push sometimes is very,

28:29 very,

28:30 let's say

28:31 good

28:33 because of course it's

28:35 pushing the market to adopt it.

28:39 The readiness,

28:40 it's of course important as I told the request to pay

28:44 can be a little bit more complicated to

28:46 be implemented than an instant credit transfer.

28:51 So that the fact that the bank and PSB are ready,

28:54 it's actually a key factor.

28:56 Standardization interoperability.

29:00 business case for merchant,

29:02 I believe that it's another key.

29:05 We need

29:05 to make the business case

29:08 very clear

29:09 to the merchant,

29:12 specifically in the pool

29:14 situation,

29:15 the payee

29:17 actually is driving that option.

29:20 So it's very important that it's clear

29:23 for the pay the

29:24 value of the request to pay.

29:26 Of course the fee model as for everything is influencing

29:30 and the trust and user experience for the for the consumer.

29:37 These are some lesson learned

29:41 trust

29:42 for specifically for the user

29:45 and also the

29:46 simplicity in some way to use it.

29:49 It's important for the user to

29:54 let's say trust this new actually

29:58 method.

29:59 Interoperability,

30:01 it's important.

30:02 I would not say that it's the most important thing because now

30:05 ISO 2022,

30:07 so

30:08 the interoperability is quite embedded in the use of it.

30:13 Regulatory clarity is fundamental,

30:15 so

30:16 that is actually,

30:17 it's one of the pillars of

30:20 making the system

30:22 be trusted by the users.

30:26 Merchant led rollout,

30:27 I already mentioned it.

30:29 The

30:31 payee

30:32 side,

30:32 it's actually

30:35 defining the success and the adoption of the request to pay

30:39 and then of course the adoption of the instant payments.

30:43 The request to pay

30:44 cannot be launched

30:46 like

30:47 in one shot

30:48 also because the request to pay is related to specific use case.

30:53 So

30:54 it can be introduced

30:56 in phases.

30:57 Actually it's a master

31:00 for the reason that every use case is different and may require different rules,

31:04 may require different integration.

31:06 So step by step starting for the most interesting use case

31:11 we are mentioning

31:13 before

31:14 utilities bills so that it's actually

31:18 probably the most common

31:20 starting point

31:22 and then of course enlarge and then also lead the market to

31:27 suggest how

31:28 request to pay can help

31:29 to create a new.

31:31 A new services,

31:33 innovative services,

31:35 it must be aligned with the national payment system strategy.

31:38 And time to market,

31:41 meaning you need to balance what the

31:43 sophistication you want to introduce and how long

31:47 you want to take to introduce in the market.

31:50 Sometime as a request to pay can be very,

31:53 very sophisticated,

31:54 but the last thing you want,

31:56 and you take 2 years to implement

31:59 use case.

32:00 So

32:01 even in a specific use case,

32:04 the single use case can be.

32:06 Let's say splitting steps,

32:09 so using

32:10 maybe

32:11 basic functionality for a specific use case and then evolve

32:14 over time.

32:19 Lesson learned last,

32:20 I think I just want to mention structured data

32:24 that they say hidden value.

32:27 What is important is that the ISO message can carry

32:32 remittance detail.

32:34 So

32:35 actually

32:37 the request to pay can carry a lot of data

32:40 inside

32:41 also depending on the implementation of the system,

32:43 but I mean.

32:45 Theoretically can include a lot of data that can of course streamline

32:50 the business processes,

32:51 so link the back office processes to the payments

32:55 to the payment leg.

32:57 So that's an automate

32:59 give at least the possibility to automate end to end

33:02 the process from the

33:04 let's say,

33:05 generation of the bill

33:07 to the payment.

33:09 Execution.

33:14 Now I will switch to confirmation of pay.

33:17 Confirmation in reality is a totally different,

33:21 is a totally different focus.

33:23 It was actually introduced

33:26 to

33:26 mitigate fraud.

33:28 Uh

33:30 That's actually,

33:32 uh,

33:32 that of course,

33:34 uh,

33:34 came in when uh instant payment system was introduced.

33:38 Reducing misdirected and erroneous payment.

33:40 Instant payment,

33:41 as you know,

33:41 it's a good thing that the funds are immediately

33:45 available to the payee

33:48 and debited to the payer.

33:50 But if I made a mistake,

33:51 there is no way back,

33:54 so.

33:55 The confirmation of payee

33:58 also

33:59 was focusing on reducing erroneous payment.

34:03 As a consequence,

34:05 they were both

34:07 these two objectives

34:10 to build trust because

34:12 of course

34:13 instant payment system

34:14 account to account payments

34:17 were introduced and of course the customers say,

34:19 OK,

34:19 instant payments is fantastic,

34:22 but

34:24 it's

34:25 it's a

34:27 it's open up of more fraud.

34:31 And

34:33 if I make an error,

34:35 it's difficult to have the money back.

34:37 So it was like these

34:41 trust issues from the users using the instant payment system.

34:45 The other thing is to improve KYC and email compliance and then we see,

34:49 because

34:50 of course the

34:52 confirmation of payee giver

34:54 can give the

34:55 certainty of who the payee is.

34:59 Also can be used to have a

35:01 more accurate KYCNML.

35:04 Compliance checks,

35:05 but

35:06 on this I will also open up a a new trend that is

35:11 that is coming on this.

35:13 This is very simple.

35:14 It is again,

35:16 we have an agreement within a payment per year on an exchange.

35:23 Then

35:24 they basically

35:26 starts before the payment

35:29 starts at a confirmation of payee request.

35:33 That is going from the pay participant to pay participant,

35:35 and the pay participant

35:37 will send back

35:40 now can be

35:42 a match,

35:43 no match,

35:43 meaning I can tell you if the

35:46 the name that you are using.

35:50 It's actually

35:51 matching,

35:52 partially matching or is no matching on actually the real name

35:56 associated to that account.

35:59 Now the payer

36:01 is presented to

36:03 more information about

36:06 the

36:07 payee.

36:08 So

36:10 I can see,

36:11 oh listen,

36:12 the name that you

36:14 you that you have

36:15 inserted actually is not the name

36:18 that is associated to the bank account.

36:20 Please check.

36:22 Uh,

36:23 maybe you make some mistakes

36:25 or maybe it's a fraud.

36:27 I means someone asks you to pay,

36:30 um,

36:31 and impersonating someone else that,

36:32 that's actually the.

36:34 There are many ways to do a confirmation of payee.

36:38 These I explained the same because

36:40 in some

36:41 jurisdictions the payee also can send back

36:45 the full name instead of saying match no match,

36:48 can send the full name

36:50 and the payer participant show the masked

36:53 name to the payer

36:55 again to support the decision.

36:57 At the end of the day,

36:58 in most jurisdictions,

36:59 the payer is in any case.

37:01 Free to start the credit transfer.

37:05 So at the end of the day,

37:06 the confirmation of pay is giving the payer

37:11 more information

37:12 to help me decide.

37:16 If you want to send this credit transfer,

37:18 the responsibility

37:20 is still on him.

37:23 to do because it's not preventing him to do it.

37:27 So that's I think also

37:29 important thing.

37:33 Core technical capabilities.

37:35 I don't want to go on all of that.

37:38 I mean I can mention a flexible transaction limit.

37:42 Maybe confirmation

37:44 of payee can be mandatory

37:46 for

37:48 above a certain threshold that can be not mandatory for small payments,

37:52 for example.

37:53 This also

37:55 actually depends on the use case.

37:57 Uh,

37:58 that you have

38:00 a confirmation of pay protocol probably um

38:04 it

38:04 it doesn't really work if you're doing a payment on a,

38:08 on a

38:08 point of sale,

38:09 let's say that.

38:11 So,

38:11 um,

38:13 let's,

38:14 depends

38:15 on also on the use case,

38:17 uh,

38:18 but it's important to that confirmation of payee can be also selective.

38:24 It's important to also to

38:27 monitor then to end

38:31 to end processing time because

38:34 now

38:36 when you do

38:37 an instant payment you also have the confirmation of pay before then.

38:41 So in the user experience from when you start the transaction to the end,

38:45 you also need the confirmation of payee to be immediate

38:50 because it influences the user experience.

38:53 And again,

38:54 of course

38:55 it's becoming

38:57 specifically where

38:58 in the jurisdiction where it's mandatory.

39:01 It of course must be available,

39:06 must have the same

39:08 requirements in terms of business continuing cybersecurity that

39:12 of the Islam Credit Council.

39:16 Models.

39:16 There are different models centralized,

39:18 decentralized hybrid centralized means that actually

39:22 everything is in a directory,

39:24 usually it's coupled with the alias directory.

39:27 That means that alias directory is also names.

39:31 So

39:32 when I ask a confirmation opinion in reality,

39:34 I am asking.

39:36 Let's say proxy resolution in terms of ISO messages,

39:38 the message is the same proxy resolution of it's called

39:42 identification identification verification

39:45 decentralized.

39:46 It's

39:48 basically.

39:49 The payer PSP sent to the payee PSP.

39:52 The message is the payee PSP that

39:54 sends back the result

39:56 of the confirmation of payee.

39:58 The hybrid that actually is not very common

40:02 where

40:03 depending on the capability of the payee PSP.

40:07 You can have both.

40:09 So

40:10 for some PSP,

40:12 it can be decentralized for pay PSP instead they can opt

40:18 not to receive the message but to opt to be to have the data in the central directory.

40:26 Some business capabilities and

40:29 of course

40:30 the possibility to manage a multiple identifier,

40:33 account proxy

40:35 LEI

40:38 response flexibility.

40:40 So

40:41 in order to have

40:42 a

40:43 rich feedback

40:44 option

40:45 here we are speaking about of course

40:48 business capability

40:51 when you can implement confirmation of payee.

40:55 This is of course

40:57 that is intrinsically was created for fraud mitigation.

41:02 And also the possibility to be cross border.

41:04 More and more confirmation of pay is

41:08 becoming a standard.

41:10 Um,

41:12 I don't want to enter.

41:14 I actually confirmation of payee in most of the use cases,

41:16 it's very important,

41:18 as I was saying.

41:19 Confirmation of payee

41:20 is starting to become

41:22 mandatory in many jurisdictions,

41:25 so you cannot do actually an instant credit transfer if there is no

41:30 confirmation of payee done

41:32 before then.

41:35 Maybe I will,

41:37 let's say mention TTR because I'm not sure if I

41:39 will be able to mention that it's a new trend.

41:42 It's

41:44 specifically when it's mandatory,

41:46 and I'm speaking

41:47 locally at also cross border

41:50 when there is some interlinking that

41:52 the confirmation of pay.

41:54 The AML and QYC checks and safety checks specifically

42:00 can be done actually at the confirmation of pay level.

42:05 So while currently

42:08 the AML checks are done

42:11 at the transaction of the payment transaction

42:13 level.

42:15 Now

42:16 if

42:17 that protocol is becoming mandatory,

42:19 you can actually move the EMH checks

42:23 at that level.

42:24 What is the advantage

42:25 that actually

42:27 you do the EML checks before starting a payment?

42:31 So if there is an issue,

42:33 you can actually

42:36 stop the transaction before

42:39 the payment,

42:39 specifically in an instant

42:41 payment.

42:42 Applying real-time AML checks.

42:46 And fraud managers check specific,

42:48 it's

42:49 a little bit more difficult

42:52 because you don't,

42:53 you have seconds to decide

42:55 if you

42:57 reject or accept the transaction

42:59 and specifically where you have

43:02 EML in EML you may have.

43:08 Let's say higher rejection rates.

43:11 So that's actually a trend that I see coming

43:15 and that's actually very interesting because the possibility to

43:18 do checks before starting a transaction where

43:23 it's much more difficult to manage

43:25 a possible AML

43:28 issue.

43:30 A

43:31 legal and economic framework,

43:34 of course it's required rule on data data occurs because actually it's all about

43:39 the accuracy of the data of the

43:41 of the users

43:44 oversight of PSP compliance and SLA that is actually important.

43:48 Privacy here we are speaking about the

43:52 the data of the of the users.

43:55 You must have a clear liability of

43:59 meaning,

44:00 as I was telling,

44:01 if

44:02 everything is done

44:03 correctly,

44:03 the liability stays on the payer side,

44:06 but if the error is created by the payee

44:10 participant,

44:12 then the liability shifts.

44:14 So if I say

44:17 a match,

44:18 but it was a false match,

44:19 now the liability is shifting,

44:21 so that

44:22 must be all true.

44:23 Now clear

44:25 economy economy.

44:26 Confirmation of peace,

44:28 it should be free because actually specifically when it's mandatory.

44:34 Because it is part of the transaction.

44:37 It's part of the instant credit transaction,

44:39 so

44:41 it's actually not logic to be to be challenged,

44:45 but specifically.

44:47 It's giving value actually.

44:49 It's in some way

44:50 making the system

44:52 save money

44:54 of all the issues that are related because it's actually reducing a lot of fraud,

44:59 a lot of erroneous payments,

45:01 so a lot of issues.

45:02 So actually it's

45:05 a value

45:09 we should see it's not a cost saving

45:11 than

45:13 let's say a service to make more fees.

45:18 These are some of the

45:20 experience that we listed.

45:23 I will let you go through them.

45:25 I don't want to

45:26 get

45:26 too long.

45:28 Benefits,

45:28 I think.

45:30 That the benefits are very clear.

45:33 I mean for direct experience here in Europe,

45:36 I think it's very nice to see

45:40 when you do a credit transfer you have a green logo

45:44 that you are sending to the right person.

45:45 There is no

45:47 and actually

45:48 for me personally it's really

45:51 a relief.

45:52 I mean you actually do a credit transfer much

45:56 more relaxed

45:58 and

46:00 And of course I would want to maybe to focus

46:03 more on the challenges

46:05 because there may be data privacy concern

46:09 specifically may

46:11 think,

46:11 but in reality

46:13 the way most of the place it's used actually

46:18 these data are not actually flowing through the confirmation of payee

46:24 and let's say that in any case.

46:26 And there are very strong scheme rules that are

46:30 covering the entire

46:32 first payment system.

46:34 For the business maybe integration costs even if in my opinion

46:38 the protocol is very,

46:39 very simple.

46:41 So yes,

46:41 there will be additional,

46:43 but differently for request to pay

46:46 where I'd say the integration could be complex.

46:48 Yes,

46:48 there are additional integration costs,

46:51 not very complicated.

46:53 Infrastructure complexity for the PSP,

46:55 I believe it's more than for the PC is adding these extra steps

47:01 and

47:02 Creating the flow for the customer

47:05 that

47:07 in real time is giving him

47:10 the result of the check.

47:12 So

47:13 yeah,

47:13 there is a little bit more complex and more resource

47:16 utilization because now for every credit transfer you have additional messages

47:22 and of course also there is additional logic.

47:26 to be implemented

47:28 liability for the regulators it is important to

47:32 understand

47:33 this liability in some cases what is happening if something goes wrong.

47:38 Sometimes,

47:39 most of the time,

47:40 let's say baseline liability is on the payer side,

47:44 but in some,

47:45 in some,

47:46 some,

47:46 let's say cases

47:48 this liability can be shifted

47:51 to the pay side.

47:55 Uh,

47:57 I see a lesson learned.

47:59 I think that,

48:00 uh,

48:00 I just want to summarize,

48:03 uh,

48:05 confirmation of P is really

48:07 adding a lot of value

48:10 on,

48:10 uh,

48:11 on domestic credit transfer.

48:14 Um,

48:16 let's say,

48:16 really,

48:18 let's say adding trust or giving more trust to the consumer and reducing

48:22 the error.

48:23 But I want to say in the cross border

48:27 it's even

48:28 more important.

48:29 The error and the complication in case of

48:33 a misdirected

48:36 payment in the cross border,

48:38 it's actually huge.

48:40 The cost to fixing

48:43 errors in a

48:45 cross border

48:47 are very high.

48:49 So actually the confirmation of pay

48:52 applied to cross border payment,

48:54 it's even more available

48:56 than applied to domestic

48:58 uh payment

49:00 so that that's I think uh.

49:03 And the other thing that I believe that

49:07 more and more

49:10 confirmation is going to become mandatory.

49:12 The value is so high

49:15 for the consumer,

49:16 the business in both domessenko's border.

49:20 That is going to become mandatory I believe in

49:24 most of jurisdictions they already have it let's say potentially

49:28 and the jurisdiction that they didn't implement it because

49:31 we also need to remember confirmation it's quite a recent

49:35 uh functionality

49:37 they are going to implement it.

49:43 Let's see.

49:44 The other is important,

49:45 and I want to

49:47 add that

49:49 of course ideally the confirmation of fee should be

49:54 using the same infrastructure

49:56 of the first payment system,

50:00 same request to pay more the confirmation of

50:01 fee because it's tightly integrated to the payment

50:05 confirmation piece is to use the same rail.

50:08 The same infrastructure of the first payment system,

50:11 although actually can be

50:13 separate.

50:14 It's not mandatory at all.

50:20 What is the

50:22 what are the policy implications

50:25 for related to the confirmation of payment request to pay?

50:28 The role of the central bank is that

50:31 they can actually operate an overlay service

50:33 to ensure that there is

50:35 a neutral

50:37 service

50:38 given to the to the PSP.

50:41 Uh,

50:42 increased complexity for the central bank,

50:44 but of course

50:45 I think it aligns with the public interest,

50:48 mandates the central bank has or

50:50 the payment system operator.

50:53 Motivation it's

50:55 clarity and liability rules,

50:56 interoperability,

50:58 so they can make these interoperable along

51:01 many rails,

51:03 more sustainable economics,

51:05 the possibility to have a minimum viable product approach,

51:10 MVP approach,

51:12 and of course the possibility to have a strong adoption,

51:16 stronger adoption and education strategies.

51:20 I think that the request to pay and confirmation of pay

51:24 are becoming an integral part of a fast payment system implementation.

51:29 I don't think that nowadays someone can imagine

51:33 to implement and introduce an instant credit

51:37 transfer system and an instant

51:39 payment system without

51:41 already

51:42 thinking

51:44 on how to

51:45 implement a request to pay a confirmation of pay.

51:48 This service,

51:49 of course.

51:50 They don't necessarily need to be

51:53 included

51:55 in the fast payment system.

51:56 Platform can be separated.

51:58 It can be.

51:58 There are service providers

52:00 that are giving requests to pay conventional pay service.

52:04 That of course that gives service to the community so it can be

52:08 separable.

52:09 What I mean is

52:11 And in the fast payment system implementation.

52:14 This must be an integral part of the strategy.

52:19 In any way,

52:22 the operator of Centralum wants to

52:24 implement it,

52:25 they must be in the strategy

52:28 and in the timeline in the project from day one because they heavily influence

52:34 the,

52:36 let's say,

52:37 the user of the Easter

52:39 payment system.

52:46 Excellent.

52:46 Thank you so much.

52:47 That was a terrific breakdown

52:49 of two,

52:50 you know,

52:51 essentially essential

52:52 uh overlay,

52:53 overlay uh products.

52:56 Um,

52:56 as a quick reminder to the audience,

52:58 all of this information,

52:59 as well as the recordings of

53:00 these webinars are available on fastpayments.worldbank.org.

53:04 Uh,

53:04 we also have another upcoming webinar and dispute resolution

53:07 mechanisms.

53:08 So thank you everyone for joining and have a wonderful day and thank you,

53:12 Claudio,

53:12 for taking the time to share,

53:14 uh,

53:14 this very important information.

53:17 Thank you.

53:18 Thanks everyone.

53:19 You have a good day.

53:20 Take care.

53:21 Thank you.

showAllTimestamps
no
transcript
Hello everyone and welcome to the World Bank's webinar series on fast payments. My name is Colin Coulter, and I'm a consultant with the World Bank based in Washington DC Project Fast, or frictionless Affordable Safe Timely Transactions, is a flagship project of the World Bank sponsored by the Gates Foundation and which aims to accelerate the adoption of fast payments in low and middle income countries. Now just a reminder that you can find all the resources that the World Bank has been producing on the fast payment work, including the recordings of these webinars under Project Fast. On a dedicated website, fastpayments.worldbank.org, today we are continuing with our series this time focusing on prominent overlay services and fast-paced payment systems, and we really have the perfect guest to lead us through that conversation, Claudio Cerrazani. Claudio's career spans 25 plus years of extensive leadership, management, and global business experience in payment systems. He is a subject matter expert on real-time payments, high value in retail payments processing, central bank digital currencies, and the clearing and settlement of securities and equities. To date, Cloudy has worked with more than 60 countries' payment system infrastructures, providing business and technical advice. He has also personally shaped the transformation of more than 30 national payment system implementations. In 2018, Claudio launched DGT Solutions designed to empower central banks, financial institutions, and FinTechs to undertake complex payment system transformation initiatives. And since then, the DGT Solutions team has supported payment systems development in more than 20 countries. Such as the UK, UAE, Saudi Arabia, Montenegro, Albania, Kosovo, North Macedonia, Maldives, Lesotho, Tajikistan, Gambia, Vietnam, Palestine, etc. The list goes on. So we really have the perfect guest today to talk us through prominent overlay services and fast payment systems. Claudio, thank you so much for joining, and the floor is yours. Thank you, Colin. Thank you everyone. Today we will focus on overlay services, so overlay services that actually are around what is the fast payment system. What are overlay services? They are, of course, value added features that are, let's say, layer on the FPS core infrastructure are completing, let's say the FPS core infrastructure. We will say we will see how and what actually, what is the scope of the various services is basically to make an instant credit transfer. And cover use cases that a normal push payment could not cover. A normal push payment with an account identification could not cover. So that actually is allowing an FPS to expand its reach and cover multiple use cases. And of course what the benefit is that of course it's promoting adoption and competition, innovation because it actually creates a layer of business functionality that can be used to create a new service and of course as a consequence of inclusion because it can reach. Any kind of a person, any kind of business. These are just a list of common overlay services. Actually, the list is increasing and you can see a request to pay, confirmation of payee, proxy directory using a flies, biller directory that actually integrating billing system with the fast payment system, tokenization. Let's say changing the account number in random tokens in order to for privacy white label application enabling the PSPs to use a pre adapted preformed white label application for management, the possibility to For the FPS operator to install a centralized management system and share the service with the PSPs, core code Man, a centralized system that can issue core codes. Also the FPS, I think as we discussed in other webinars, can also be leveraged to do an open banking app now today. We will just focus on the, let's say, what is the prominent of services that are requested to pay a confirmation of pay. There are general capabilities that are common to these two overlay services request to pay and confirmation of the Messaging standards. Currently ISO 2022, it's dominant. It's of course it's de facto standard. Some of jurisdictions are still not using ISO, but it's more an historical reason. And let's say all the most recent implementation and of course the future implementation will adopt ISO 20022, very flexible and with specific messages for requests to pay and confirmation of pay. Operational hours, these are the service of a fast payment system and of course they must follow the same availability standards, so they must run 24/7. API more and more overlay services are based on API more than messaging. So and we will see that let's say that the trend in the market is that specifically for the services. And the use of API is becoming predominant. The use capability, the possibility to use these services using aliases, so both a request to pay and confirmation of pay as the possibility to use aliases instead account identifications. Access channels. These services must be, must be available on all channels, same as fast payment system participation. Now participation, it's another. Let's say specific um. Uh, of, uh, the, the confirmation of PNR requests to pay that actually can have uh It can be treated as a separate service to the first payment, although they are complementing the fast payment system and they are enabling fast payment system use cases, they can also be seen as a separate service, so the participation to this overlay service can also be separated. Let's say that participant to overlay service can be more or less than the participant of the FPS. Business continuity, of course, exactly this is because it's coupled with a fast payment transaction. You must take care of the business continuity with the same care as the fast payment system. Now, first, let's start with request to pay. What is the goal of request to pay? Streamline payments, enhance flexibility, improve cash flow, and expand FPS adoption, um. Motivation are many and let's say you can see it here in the slides, but what I want to also to stress that the request to pay is A really powerful tool. Request to pay protocol. It's extremely flexible, so it can really be used also beyond the use case that I will now show because actually it's open to creativity from the PSP also to create new services. These are let's say basic request to pay transaction flow. There is of course an engagement with the payer and the payee on buying goods, services, and point number 1 when agreed this exchange, the payee can send a request to pay to the payer. It's sometimes also called the pool. Payment. So the payee is pulling the payment. The payer can offer accept, reject, and then we will see there are also other menu options depending on the use case. And let's say if you accept now the step 4, there is the credit transfer that starts as a normal, let's say credit transfer, instant credit transfer. Uh, these are some capabilities that should be taken into consideration where implementing a request to pay now uh just in a Let's say in terms of experience of implementation, um, not the request to pay a different maturity and then of course not to request to pay. As all the capabilities that we can see here. This is more a suggestion for a way forward. These are also including sophisticated capabilities that are enabling a more sophisticated use case on request to pay. Also, request to pay can be very simple. Anyway, I just mentioned, of course, the messaging as it was pretty much standard. Pane 1314 is the ASU standard. One thing that it's important in SLA management, so the capability to be sure that the end to end transaction, both the request to pay and the linked payment are enforced, are enforced by the system. The idea is that the request to pay can support multiple instruments. Another thing that it's we are speaking about first payment, but the request to pay as a tool can be also seen as a separate service. So a request to pay can also trigger a payment that actually is not a fast payment. So in some implementation, the request to pay, it's actually A wider than the first payment itself. So after request to pay, it's not mandatory to send an instant payment, but you can also complete with different instruments. So it can be connected or not. Let's say for the maximum flexibility it's better to keep it open. And the possibility to monitor the processing time in the different steps and legs and of course the possibility to use the same infrastructure to handle both the payment trail and the new financial message as also the. A request to pay and confirmation of the app code, so payment, nonpayment messages, sometimes financial, nonfinancial messages. Fraud, we need to also understand that the request to pay needs also fraud prevention because it's a pool payment and being a pool payment it's also open to let's say the same type of fraud that an instant credit transfer may have. So important focus on fraud prevention and there are multiple ways. Transaction limits, and I think it's also to define the possibility to define. Also rules that are that are related to transaction limits, how a request to pay can be used in different use cases, what are, what are the limits you can put depending on where the request to pay is used. Retailer measure is different from a request to pay using business to business or in a billing situation. Of course, another thing that it's important is a granular message status manager, so the possibility to categorize the request to pay in pending paid, not paid, rejected to be able to actually track the life cycle of requests to pay also connected to the related payments there can be one or multiple. And these are some of the business capabilities that are important, so the possibility to have a single request to pay, a recurring request to pay, also a bulk request to pay. You can see that in some government or bill application, having the possibility to have a bulk request to pay can be important. The possibility to have a flexible acceptance option. So for the payer accept now or accept later. Possi to have the flexible payment mode. Pay now. I can pay later. I can make partial payments, pay some of what was requested. And so these are the business capabilities that are very, very interesting if you implement a request to pay. But if you get a cancellation, the possibility also to cancel a request to pay in certain use cases when your request to pay is used to agree, let's say when a request to pay is used for a similar direct debit. You need also to be able to cancel this agreement that was taken with a request to pay, and of course you can also use the request to pay for other specific. Meaning you can have multiple requests to pay to agree between payer and payee. You can use the request to pay to send reminders of payment also to trigger OTP or approval to give extra security. I'm speaking about this, but there are actually, it's not actually theory. They put in practice all these business capabilities. Make the RTP much very, very powerful. These are some use cases as I told the use case that the request to pay can cover are many here I'm just, let's say listing the basic ones that is one of FTP that it's actually the pay and FTP you accept reject, and basically it's giving Full control to the payer, uh, very, very transparent, uh, and can be even real-time, um. It's, it's an idea for adopt payments or govern to payment govern to personal outreach, and it's actually one of the basic use cases that let's say that an RTP can cover. Then we have the recurrent RTP that actually probably is the most interesting use of it. Utilities, rent, bills, telecom billing subscription, membership insurance, it's very powerful. It can of course actually be a valid alternative to the direct debit. Actually it's much more flexible than a direct debit, and it is also can be also more transparent for the for the payer. So it can also enhance. The way direct debit is done. And create a lot of flavors, let's say, of let's say direct debits. Then there is the last that I want to mention point of interaction request to pay. Why it's important, it's because a request to pay if used on a POS can mimic a typical card transaction. Usually when I mean when you want to go on the merchant, the most common way it's the payer, the merchant started the transaction, so the payer exchanged, insert the card, exchange some data with NFC. And the merchant starts the transaction. Of course in a in a fast payment system that is a push payment, you typically the basic use case on the merchant side is actually the opposite. The merchant show a core code, I read the core code, and I push a transaction. Using the request to pay, in reality, unlock more use case that can be done with an instant payment, instant payment system, an instant credit transfer, because with the request to pay, for example, let's say that I have my, now I'm making it simple, but actually it's much more complex, but if I have a card with my account ID and I tap on the POS. the POS, actually the merchant start a request to pay. That goes to my bank. The bank basically can there are some, let's say, rules. Consider that a pre-approved transaction if you have the money in the account, goes to the confirmation and the merchant, of course, gives you the goods. So that's actually now speak just one of the cases, but give the idea on what request to pay can unlock when used on a on a point of interaction. Uh, Claudio, this is, I really appreciate the breakdown on the kind of specific use cases. I don't want to slow us down, but I'm curious because in earlier slides, you're talking about the importance of how this can help build trust in the payment system, how the importance of like fraud protection stuff, um, of course, wrapped up within an overlay services program. I'm curious, back on that slide, are any of these use cases. Do you think they are maybe more promising in building trust when people are first launching one of these systems than any of the others? I'm just thinking about it, it's fully open. Everyone can kind of use this. If you're building an FPS launching or launching an RTP system, which of these is kind of the strongest way to build trust in users? OK, every, every kind of services must be introduced with care because it gives you the potential to have new use case, but it's also opens up to new threats. And request to pay you can may understand if you, for example, if you allow everyone to send a request to pay to everyone else and you may be receiving a request to pay from you don't know and you need to before introducing a request to pay, you need to analyze the specific use case and put Very specific rules on it in order to create the trust. If I would like to, let's say start creating trust, I would start with, let's say more utilities, more like business to person, where in reality, of course it's more controlled because you can control who is generating the request to pay. And a lot and you actually give more transparency and more data to the payer. And to check what he has to pay, so I send you an invoice. I send you a subscription. I can see, I can check it's good that it's what I was expecting, and I can pay immediately actually in the same in the same transaction or later. I mean it depends probably a utilities bill, it would be have a pay later functionality, you know, I want to check and I can pay. And I can probably sometime put a time. I mean, OK, I agree I pay at this date. The sophistication of request to pay and the flexibility, it really can be very, very high and give you all these, and I think that it's what a customer will like most, and you are in a safe situation, then little by little can expand on other use cases. Thank you. That's very helpful. It's almost like the, the real value creation is that transparency that then builds the trust, choice. Uh, thank you. I don't want to slow us down. OK, now some legal regulatory economic aspect aspects on the legal and regulatory, of course it's important to specify when a request to pay is a binding obligation, meaning when, and that is of course it's done with and enforced with request to pay scheme rules so very clear. Um, very, very, very clear definition of the obligation of the participants in this overlay service. Then of course also Ensuring customer potential, so a very clear dispute resolution mechanism, very clear liabilities. between the participant who is paying for if something goes wrong and then and that this can also be different depending on the use case that you are implementing. If the request to pay is implemented in a point of interaction, you can have different obligation, different dispute resolution mechanisms than if you are doing, let's say, a request to pay for a utility bill. In reality, every use case most probably would have its own rules. In request to pay, you don't actually have a request to pay as a protocol and scheme rules of request to pay. Actually, request to pay, it's very different in the way it's used. I mean if I use a use case it can be a totally different business process and obligation than in another use case that must be very, very clear. So it can be integrated in the fast payment system rule. It's usually suggested to be separate exactly for the reason I was just saying because the the RTP rule can be actually quite complicated and quite different depending on use case to use case. In terms of fee models, actually there are many, many models, but usually because the request to pay is triggering a fast payment system, a fast payment transaction, sorry. And you usually, let's say the approach is to make it free or very low in order to create more adoption because actually it's pulling more transaction on an instant payment system and again depending on the use case on a P2P most probably you should go for no no fee. Of course in a business situation you may also let the participant, let's say be more free or charge a higher fee. So that's also another thing. The fee model can vary. On a specific use case where they request to pay is used. Let's see the timing. OK, I don't want to too much on the request to pay benefit the challenges, uh, I believe, uh, every request to pay has a lot of benefits that I can, I can really highlight that. But of course, again, this new service must be introduced with care because depending on the use case, depending on is implemented. You may have issues of privacy. You may issue of fraud. Sometimes you have, you can also confuse the payer. So the other thing for for business for PSP can be complex depending on how the protocol is created. So you may have additional cost. Liability can be different. So also the business process that you need to put in place can be complicated. You need a specific, let's say, a regulation or rules around it that must be followed. And you need to have clearer legal liability rules that again when you think of instant payment system, it's one rule when clarify that and the request to pay can be actually multiple depending on how you use it. I mean here we just put some kind of example of request to pay. In reality, despite the power of request to pay, the user request to pay, it's not so wide as it should be in my opinion, and I think it's because of Let's say some sometimes being having something very powerful makes it difficult to let's say apply it to a specific case and break down all the business process, understand the make the consumer understand the value and the businesses. Anyway, let's say that the request to pay, if you can see, was launched in the most prominent region and jurisdictions. So it's something that it's in my opinion, a key for a fast payment system to evolve. These are some of the cross-cutting success factors. Regulatory push sometimes is very, very, let's say good because of course it's pushing the market to adopt it. The readiness, it's of course important as I told the request to pay can be a little bit more complicated to be implemented than an instant credit transfer. So that the fact that the bank and PSB are ready, it's actually a key factor. Standardization interoperability. business case for merchant, I believe that it's another key. We need to make the business case very clear to the merchant, specifically in the pool situation, the payee actually is driving that option. So it's very important that it's clear for the pay the value of the request to pay. Of course the fee model as for everything is influencing and the trust and user experience for the for the consumer. These are some lesson learned trust for specifically for the user and also the simplicity in some way to use it. It's important for the user to let's say trust this new actually method. Interoperability, it's important. I would not say that it's the most important thing because now ISO 2022, so the interoperability is quite embedded in the use of it. Regulatory clarity is fundamental, so that is actually, it's one of the pillars of making the system be trusted by the users. Merchant led rollout, I already mentioned it. The payee side, it's actually defining the success and the adoption of the request to pay and then of course the adoption of the instant payments. The request to pay cannot be launched like in one shot also because the request to pay is related to specific use case. So it can be introduced in phases. Actually it's a master for the reason that every use case is different and may require different rules, may require different integration. So step by step starting for the most interesting use case we are mentioning before utilities bills so that it's actually probably the most common starting point and then of course enlarge and then also lead the market to suggest how request to pay can help to create a new. A new services, innovative services, it must be aligned with the national payment system strategy. And time to market, meaning you need to balance what the sophistication you want to introduce and how long you want to take to introduce in the market. Sometime as a request to pay can be very, very sophisticated, but the last thing you want, and you take 2 years to implement use case. So even in a specific use case, the single use case can be. Let's say splitting steps, so using maybe basic functionality for a specific use case and then evolve over time. Lesson learned last, I think I just want to mention structured data that they say hidden value. What is important is that the ISO message can carry remittance detail. So actually the request to pay can carry a lot of data inside also depending on the implementation of the system, but I mean. Theoretically can include a lot of data that can of course streamline the business processes, so link the back office processes to the payments to the payment leg. So that's an automate give at least the possibility to automate end to end the process from the let's say, generation of the bill to the payment. Execution. Now I will switch to confirmation of pay. Confirmation in reality is a totally different, is a totally different focus. It was actually introduced to mitigate fraud. Uh That's actually, uh, that of course, uh, came in when uh instant payment system was introduced. Reducing misdirected and erroneous payment. Instant payment, as you know, it's a good thing that the funds are immediately available to the payee and debited to the payer. But if I made a mistake, there is no way back, so. The confirmation of payee also was focusing on reducing erroneous payment. As a consequence, they were both these two objectives to build trust because of course instant payment system account to account payments were introduced and of course the customers say, OK, instant payments is fantastic, but it's it's a it's open up of more fraud. And if I make an error, it's difficult to have the money back. So it was like these trust issues from the users using the instant payment system. The other thing is to improve KYC and email compliance and then we see, because of course the confirmation of payee giver can give the certainty of who the payee is. Also can be used to have a more accurate KYCNML. Compliance checks, but on this I will also open up a a new trend that is that is coming on this. This is very simple. It is again, we have an agreement within a payment per year on an exchange. Then they basically starts before the payment starts at a confirmation of payee request. That is going from the pay participant to pay participant, and the pay participant will send back now can be a match, no match, meaning I can tell you if the the name that you are using. It's actually matching, partially matching or is no matching on actually the real name associated to that account. Now the payer is presented to more information about the payee. So I can see, oh listen, the name that you you that you have inserted actually is not the name that is associated to the bank account. Please check. Uh, maybe you make some mistakes or maybe it's a fraud. I means someone asks you to pay, um, and impersonating someone else that, that's actually the. There are many ways to do a confirmation of payee. These I explained the same because in some jurisdictions the payee also can send back the full name instead of saying match no match, can send the full name and the payer participant show the masked name to the payer again to support the decision. At the end of the day, in most jurisdictions, the payer is in any case. Free to start the credit transfer. So at the end of the day, the confirmation of pay is giving the payer more information to help me decide. If you want to send this credit transfer, the responsibility is still on him. to do because it's not preventing him to do it. So that's I think also important thing. Core technical capabilities. I don't want to go on all of that. I mean I can mention a flexible transaction limit. Maybe confirmation of payee can be mandatory for above a certain threshold that can be not mandatory for small payments, for example. This also actually depends on the use case. Uh, that you have a confirmation of pay protocol probably um it it doesn't really work if you're doing a payment on a, on a point of sale, let's say that. So, um, let's, depends on also on the use case, uh, but it's important to that confirmation of payee can be also selective. It's important to also to monitor then to end to end processing time because now when you do an instant payment you also have the confirmation of pay before then. So in the user experience from when you start the transaction to the end, you also need the confirmation of payee to be immediate because it influences the user experience. And again, of course it's becoming specifically where in the jurisdiction where it's mandatory. It of course must be available, must have the same requirements in terms of business continuing cybersecurity that of the Islam Credit Council. Models. There are different models centralized, decentralized hybrid centralized means that actually everything is in a directory, usually it's coupled with the alias directory. That means that alias directory is also names. So when I ask a confirmation opinion in reality, I am asking. Let's say proxy resolution in terms of ISO messages, the message is the same proxy resolution of it's called identification identification verification decentralized. It's basically. The payer PSP sent to the payee PSP. The message is the payee PSP that sends back the result of the confirmation of payee. The hybrid that actually is not very common where depending on the capability of the payee PSP. You can have both. So for some PSP, it can be decentralized for pay PSP instead they can opt not to receive the message but to opt to be to have the data in the central directory. Some business capabilities and of course the possibility to manage a multiple identifier, account proxy LEI response flexibility. So in order to have a rich feedback option here we are speaking about of course business capability when you can implement confirmation of payee. This is of course that is intrinsically was created for fraud mitigation. And also the possibility to be cross border. More and more confirmation of pay is becoming a standard. Um, I don't want to enter. I actually confirmation of payee in most of the use cases, it's very important, as I was saying. Confirmation of payee is starting to become mandatory in many jurisdictions, so you cannot do actually an instant credit transfer if there is no confirmation of payee done before then. Maybe I will, let's say mention TTR because I'm not sure if I will be able to mention that it's a new trend. It's specifically when it's mandatory, and I'm speaking locally at also cross border when there is some interlinking that the confirmation of pay. The AML and QYC checks and safety checks specifically can be done actually at the confirmation of pay level. So while currently the AML checks are done at the transaction of the payment transaction level. Now if that protocol is becoming mandatory, you can actually move the EMH checks at that level. What is the advantage that actually you do the EML checks before starting a payment? So if there is an issue, you can actually stop the transaction before the payment, specifically in an instant payment. Applying real-time AML checks. And fraud managers check specific, it's a little bit more difficult because you don't, you have seconds to decide if you reject or accept the transaction and specifically where you have EML in EML you may have. Let's say higher rejection rates. So that's actually a trend that I see coming and that's actually very interesting because the possibility to do checks before starting a transaction where it's much more difficult to manage a possible AML issue. A legal and economic framework, of course it's required rule on data data occurs because actually it's all about the accuracy of the data of the of the users oversight of PSP compliance and SLA that is actually important. Privacy here we are speaking about the the data of the of the users. You must have a clear liability of meaning, as I was telling, if everything is done correctly, the liability stays on the payer side, but if the error is created by the payee participant, then the liability shifts. So if I say a match, but it was a false match, now the liability is shifting, so that must be all true. Now clear economy economy. Confirmation of peace, it should be free because actually specifically when it's mandatory. Because it is part of the transaction. It's part of the instant credit transaction, so it's actually not logic to be to be challenged, but specifically. It's giving value actually. It's in some way making the system save money of all the issues that are related because it's actually reducing a lot of fraud, a lot of erroneous payments, so a lot of issues. So actually it's a value we should see it's not a cost saving than let's say a service to make more fees. These are some of the experience that we listed. I will let you go through them. I don't want to get too long. Benefits, I think. That the benefits are very clear. I mean for direct experience here in Europe, I think it's very nice to see when you do a credit transfer you have a green logo that you are sending to the right person. There is no and actually for me personally it's really a relief. I mean you actually do a credit transfer much more relaxed and And of course I would want to maybe to focus more on the challenges because there may be data privacy concern specifically may think, but in reality the way most of the place it's used actually these data are not actually flowing through the confirmation of payee and let's say that in any case. And there are very strong scheme rules that are covering the entire first payment system. For the business maybe integration costs even if in my opinion the protocol is very, very simple. So yes, there will be additional, but differently for request to pay where I'd say the integration could be complex. Yes, there are additional integration costs, not very complicated. Infrastructure complexity for the PSP, I believe it's more than for the PC is adding these extra steps and Creating the flow for the customer that in real time is giving him the result of the check. So yeah, there is a little bit more complex and more resource utilization because now for every credit transfer you have additional messages and of course also there is additional logic. to be implemented liability for the regulators it is important to understand this liability in some cases what is happening if something goes wrong. Sometimes, most of the time, let's say baseline liability is on the payer side, but in some, in some, some, let's say cases this liability can be shifted to the pay side. Uh, I see a lesson learned. I think that, uh, I just want to summarize, uh, confirmation of P is really adding a lot of value on, uh, on domestic credit transfer. Um, let's say, really, let's say adding trust or giving more trust to the consumer and reducing the error. But I want to say in the cross border it's even more important. The error and the complication in case of a misdirected payment in the cross border, it's actually huge. The cost to fixing errors in a cross border are very high. So actually the confirmation of pay applied to cross border payment, it's even more available than applied to domestic uh payment so that that's I think uh. And the other thing that I believe that more and more confirmation is going to become mandatory. The value is so high for the consumer, the business in both domessenko's border. That is going to become mandatory I believe in most of jurisdictions they already have it let's say potentially and the jurisdiction that they didn't implement it because we also need to remember confirmation it's quite a recent uh functionality they are going to implement it. Let's see. The other is important, and I want to add that of course ideally the confirmation of fee should be using the same infrastructure of the first payment system, same request to pay more the confirmation of fee because it's tightly integrated to the payment confirmation piece is to use the same rail. The same infrastructure of the first payment system, although actually can be separate. It's not mandatory at all. What is the what are the policy implications for related to the confirmation of payment request to pay? The role of the central bank is that they can actually operate an overlay service to ensure that there is a neutral service given to the to the PSP. Uh, increased complexity for the central bank, but of course I think it aligns with the public interest, mandates the central bank has or the payment system operator. Motivation it's clarity and liability rules, interoperability, so they can make these interoperable along many rails, more sustainable economics, the possibility to have a minimum viable product approach, MVP approach, and of course the possibility to have a strong adoption, stronger adoption and education strategies. I think that the request to pay and confirmation of pay are becoming an integral part of a fast payment system implementation. I don't think that nowadays someone can imagine to implement and introduce an instant credit transfer system and an instant payment system without already thinking on how to implement a request to pay a confirmation of pay. This service, of course. They don't necessarily need to be included in the fast payment system. Platform can be separated. It can be. There are service providers that are giving requests to pay conventional pay service. That of course that gives service to the community so it can be separable. What I mean is And in the fast payment system implementation. This must be an integral part of the strategy. In any way, the operator of Centralum wants to implement it, they must be in the strategy and in the timeline in the project from day one because they heavily influence the, let's say, the user of the Easter payment system. Excellent. Thank you so much. That was a terrific breakdown of two, you know, essentially essential uh overlay, overlay uh products. Um, as a quick reminder to the audience, all of this information, as well as the recordings of these webinars are available on fastpayments.worldbank.org. Uh, we also have another upcoming webinar and dispute resolution mechanisms. So thank you everyone for joining and have a wonderful day and thank you, Claudio, for taking the time to share, uh, this very important information. Thank you. Thanks everyone. You have a good day. Take care. Thank you.
showAllTranscripts
no
duration
PT53M24S
scene7File
worldbank/Overlay-Services-Webinar-Recording-20260128-080546
scene7Domain
https://worldbank.scene7.com/
scene7FileAvs
worldbank/Overlay-Services-Webinar-Recording-20260128-080546-AVS
title
Overlay-Services-Webinar-Recording-20260128-080546
description
Overlay-Services-Webinar-Recording-20260128-080546
showTimestampAndTranscript
yes
col-xs-12
col-sm-12
col-md-3
col-lg-3
col-xs-12
col-sm-12
col-md-7
col-lg-7
  • add-style
  • lp-body-content

Also known as instant or real-time payments, fast payments are characterized by the instant transmission of the payment message and by the immediate availability of funds to the beneficiary on a 24/7/365 basis. But authorizing, clearing, and settling a payment are but one layer of a robust fast payment system. Atop them lies a variety of overlay services that can improve product value, prevent fraud, broaden adoption, and drive the overall success of the fast payment system.

In this webinar, Colin Colter (Payments Consultant, World Bank) and Claudio Ceresani (Senior Payments Consultant, World Bank) discuss the role of overlay services (in particular request-to-pay and confirmation of payee) in fast payment systems, how they help build trust and adoption of an FPS, and how to operationalize them at the system level.

lp-heading-top-medium
lp-heading-bottom-medium
col-xs-12
col-sm-12
col-md-2
col-lg-2