ANALYSIS OF BANKS’ RESPONSES TO THE QUESTIONNAIRE ON ... · Questionnaire, regarding the implementation of measures for the ML/TF risk management and control in the period from

AML SUPERVISION CENTRE

ANALYSIS OF BANKS’ RESPONSES TO THE

QUESTIONNAIRE ON ACTIVITIES REGARDING

ML/TF RISK MANAGEMENT FOR Q1 2019

Belgrade, July 2019

Analysis of Banks’ Responses to the Questionnaire on Activities Regarding ML/TF Risk Management for Q1 2019 National Bank of Serbia

1

Contents:

I Key findings 3

II General bank data 4

III Clients 5

IV Client composition by the assessed risk level 7

V Client composition by CDD actions and measures 9

VI Transactions and products 10

VII Outsourcing CDD actions and measures to third parties 13

VIII Correspondent relationship 13

IX Corporative governance 13

X Employee training 14

XI Organisational structure 15

XII Reporting to the AML/CFT compliance officer 15

XIII Internal audit and internal control 16


2

Introductory note

Within bank supervision, the National Bank of Serbia carries out off-site monitoring and analysis of the ML/TF risk management activities of banks. The analysis of data from the Questionnaire on Banks’ Activities regarding ML/TF Risk Management (hereinafter: Questionnaire) is carried out at least once a year. The National Bank of Serbia has analysed the Questionnaire since 2006, which has been changed and adjusted as a consequence of the adoption of the Law on the Prevention of Money Laundering and Terrorism Financing (hereinafter: the Law) and the NBS Decision on the Guidelines for Assessing the Risk of Money Laundering and Terrorism Financing (RS Official Gazette, Nos 13/2018 an 103/2018). After the National Money Laundering and Terrorism Financing Risk Assessment was carried out in 2018, a new Questionnaire was created and updated for the purposes of further improving the risk-based approach to supervising banks’ activities relating to the ML/TF risk management.

Questions in the Questionnaire are sorted into twelve groups:

Part I: General bank data

Part II: Clients

Part III: Client composition by the assessed risk level

Part IV: Client composition by CDD actions and measures

Part V: Transactions and products

Part VI: Outsourcing CDD actions and measures to third parties

Part VII: Correspondent relationship

Part VIII: Corporate governance (in the part of the information system relating to the ML/TF risk)

Part IX: Employee training

Part X: Organisational structure

Part XI: Reporting to the AML/CFT compliance officer

Part XII: Internal audit and internal control

The key objectives of the analysis of data from the Questionnaire: analysing the state-of-play in terms of identifying, measuring and managing the ML/TF risk in the entire banking system for the purposes of timely identification of areas that may indicate increased exposure to the ML/TF risk; off-site monitoring of the effectiveness and adequacy of the established ML/TF risk management system in the entire banking system and identification of possible deficiencies in the risk management system; timely alerting banks to potential exposure to the ML/TF risk.

This analysis was carried out based on data for January–March 2019, submitted by banks to the NBS. The presented data may, therefore, exhibit


3

certain inconsistencies, and are mostly analytical in nature. In the period covered by the analysis, 27 banks operated in Serbia’s banking sector.

I Key findings

Based on the analysis, banks have medium exposure to the ML/TF risk, as in the previously analysed periods. Banks are exposed to the ML/TF risk due to the size of the sector and the branched out network, the number and type of clients, as well as the number of performed transactions.

As at 31 March 2019, 27 banks established business relations with clients. Compared to the previously analysed period, there was a negligible change in the number of locations at which a bank could establish business relations with clients and locations at which cash transactions could be executed. The total number of clients on the said date was 11,953,946, of which 11,169,401 natural persons (76,787 non-residents), 395,921 legal persons (5,225 non-resident legal persons) and 388,624 entrepreneurs. In the observed period, over 107 million cashless and over 33 million cash transactions were executed.

The key findings based on the analysis of the data provided in the Questionnaire, regarding the implementation of measures for the ML/TF risk management and control in the period from 1 January to 31 March 2019 are, inter alia, the following:

- All banks set up their own ML/TF risk management systems applying an ML/TF risk assessment approach and taking into account the findings of the National Money Laundering Risk Assessment;

- 26 out of 27 banks use some of the software for tracking transactions and clients to detect suspicious transactions and individuals, while one bank is in the process of introducing the software;

- All 27 banks stated they use Lists of indicators for detecting suspicious transactions regarding money laundering, indicators from the List of indicators for detecting suspicious transactions regarding terrorism financing, published on the website of the Administration for the Prevention of Money Laundering (hereinafter: Administration), as well as lists of designated persons, published by the UN Security Council and other international organisations of which Serbia is a member;

- 13 banks terminated business relations with 36,550 clients because it was impossible to carry out CDD actions and measures in the period from 1 January until 31 March 2019;

- 15 banks refused the offer to establish a business relationship and/or execute a transaction in 8,247 cases, due to the impossibility to take CDD actions and measures in the period from 1 January until 31 March 2019;


4

- In the overall banking sector, 8 banks used the possibility to entrust some of the CDD actions and measures to third parties in accordance with the Law;

- Over the past year, 25 banks carried out AML/CFT-related audit, and two banks, which failed to do so, carried out their audits in 2015 and 2016. The audit identified deficiencies in 23 banks and set deadlines for their removal.

- Over the past year, 26 banks carried out AML/CFT-related internal control, and the one bank, which failed to do so, carried out its internal control in Q3 and Q4 2017. Internal controls identified deficiencies in 25 banks and set deadlines for their removal;

- All banks stated they introduced their employees with the consequences of non-compliance with laws, procedures and findings of the ML/TF risk management control. As at 31 March 2019, 11,631 employees completed the relevant training successfully.

II General bank data

As at 31 March 2019, there were 27 banks licensed by the NBS in the Republic of Serbia. According to the submitted responses, banks established business relations with clients at 2,313 locations within the business network of the entire banking sector consisting of 745 branches, 424 branch offices, 13 regional centres and three offices for housing loans. Six banks can establish business relationships at up to 10 locations, six banks at 11–50 locations, eight banks at 51–100 locations, five banks at 101–200 locations, and two banks at over 200 locations. At banking sector level, cash transactions can be made at 4,381 locations. Six banks have up to 10 such locations, five banks between 11 and 50 locations, ten banks between 51 and 100, two banks between 101 and 200, and two banks over 200 such locations.

Total banking sector employment as at 31 March 2019 was 22,712 persons, while the number of employees directly engaged in client- and transaction-related operations (front-office) was 13,249 (58.33% of the total number of employees). The analysis of employment data reveals a continuation of the staff reduction trend, with total employment at end-March 2019 down by 157 persons relative to end-December 2018. At banking sector level, the average number of clients per employee directly engaged in client- and transaction-related operations was 902. The average number of clients per employee fell by 28 relative to 31 December 2018.


5

III Clients

As at 31 March 2019, the total number of clients at the banking sector level came at 11,953,946, down by 516,962 in absolute amount relative to the previously analysed period (as at 31 December 2017).

Based on the data obtained from the Questionnaire, Charts 1–6 show the composition of clients by residence, the legal form of organising resident legal persons, as well as the composition of non-residents in the banking sector by their country risk profile.

Relative to the previously analysed period, there were no significant changes in the composition of clients by residence in the banking sector. The total number of residents was 11,871,934 (99.31% of total clients) and of non-residents 82,012 (0.69%).

Of the total number of resident clients, natural persons remained dominant (94%), while domestic legal persons and entrepreneurs accounted for 3%, respectively (Chart 2).

Chart 1 Clients by residence as at 31 March 2019

Residents 11,871,934 - 99.31% Non-residents 82,012 - 0.68%

Chart 2 Composition of resident clients as at 31 March 2019

Natural persons - 11,092,614 - 93.43% Entrepreneurs - 388,624 - 3.27%

Legal persons - 390,696 - 3.29%


6

Relative to the previously analysed period, the composition of resident clients remained almost unchanged, with natural persons accounting for around 93.44%, entrepreneurs around 3.27% and legal persons around 3.29%.

Chart 3 shows the composition of resident legal persons by legal form of organising.

In the total number of legal persons, the dominant share belongs to limited liability companies (DOO) with 73.51%, while joint-stock companies (AD), legal persons organised in other legal forms (limited partnership, general partnership) and all other forms (associations, etc.) accounted for 2.65%, 1.44% and 22.40%, respectively.

The total number of non-resident clients in the analysed period was 82,012. Relative to the previously analysed period (as at 31 December 2018), the number of non-resident natural persons decreased by 4,006, while the number of non-resident legal persons increased by 30.

Chart 3 Composition of resident legal persons by legal form of organisation as at 31 March 2019

Limited Liability Companies - 287,210 - 73.51%

Joint Stock Companies - 10,343 - 2.64%

Other legal forms (partnerships, limited partnerships and similar ) - 5,643 - 1.44%

Remaining forms (associations and similar) - 87,500 - 22.39%

Chart 4 Composition of non-resident clients as at 31 March 2019

Natural persons - 876,787 - 93.96% Legal persons - 5,225 - 6.43%


7

The share of natural persons from high-risk countries in the total number of non-resident natural persons was 26.63%, while 5.86% of them were from the countries with strategic deficiencies in the AML/CFT system. In addition, 2.10% of the share relates to non-resident natural persons from offshore countries, while natural persons from other countries which do not have a high or elevated ML/TF risk hold a dominant 65.4% share.

As with non-resident natural persons, the dominant share in the composition of non-resident legal persons belongs to legal persons from other countries (71.46%), followed by high-risk countries (15.33%), offshore countries (12.45%) and countries with strategic deficiencies in the AML/CFT system (0.75%).

IV Client composition by the assessed risk level

In accordance with the Law and the NBS Decision on the Guidelines for Assessing the Risk of Money Laundering and Terrorism Financing, all banks have an obligation to prepare the risk analysis for each group or type of client,

Chart 6 Composition of non-resident legal persons by country risk profile as at 31 March 2019

High-risk countries - 801 - 15.33%

Countries with strategic deficiencies - 39 - 0.74%

Offshore countries - 651 - 12.45%

Other countries - 3,734 - 71.46%


8

business relation, and/or services offered by the obligor within their activities and/or transactions, taking into account the results of the National Money Laundering Risk Assessment.

Depending on the results of the analysis, banks classified their clients into the following risk categories: low-, medium- and high-risk (Chart 7).

As for the composition of high-risk clients, natural persons have the largest share (38%), followed by resident legal persons (29%), non-resident natural persons (around 22%), entrepreneurs (8%) and non-resident legal persons (around 3%), as shown in Chart 8.

Chart 9 gives an overview of the number of clients on which in the last three years banks submitted data to the Administration, as this is one of the criteria for assessing clients as high-risk.

Chart 7 Clients by risk level as at 31 March 2019

Low 5,782,776 - 48.37% Medium - 6,081,324 - 50.87% High - 89,856 - 0.70%

Chart 8 Composition of high-risk clients as at 31 March 2019

Natural persons - residents - 34,263 - 38.13%

Entrepreneurs - 7,535 - 8.38%

Legal persons - residents - 26,180 - 29.13%

Natural persons - non-residents - 119,436 - 21.63%

Legal persons - non-residents - 2,442 - 2.71%


9

The analysis has shown that most of the clients for which data were submitted to the Administration were the clients of banks occupying leading positions by the number of clients in the banking sector.

V Client composition by CDD actions and measures

Pursuant to the Law, banks are required to apply general, simplified or enhanced CDD actions and measures. In the period 1 January–31 March 2019, general actions and measures were taken in respect of 83.37% clients at banking sector level. Simplified actions and measures were taken in respect of 15.76% and enhanced in respect of 0.86% of clients (Chart 10).

The table below shows a comparative overview of the number of clients by risk level and CDD actions and measures as at 31 March 2019.

986

1,315

3,051

3,360

0

700

19963

730

1,147

655

65

2,578

6237

509

54

1,569

1,0711,367

772896

25

511

12671,315

420

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000B

ank

1

Ban

k 2

Ban

k 3

Ban

k 4

Ban

k 5

Ban

k 6

Ban

k 7

Ban

k 8

Ban

k 9

Ban

k 1

0

Ban

k 1

1

Ban

k 1

2

Ban

k 1

3

Ban

k 1

4

Ban

k 1

5

Ban

k 1

6

Ban

k 1

7

Ban

k 1

8

Ban

k 1

9

Ban

k 2

0

Ban

k 21

Ban

k 2

2

Ban

k 23

Ban

k 24

Ban

k 2

5

Ban

k 2

6

Ban

k 27

Chart 9 Number of clients whose data have been submitted to the APML in the last three years, by bank

Chart 10 Composition of clients by CDD actions and measures applied as at 31 March 2019

Simplified 1,884,575 - 15.76% General 9,966,214 - 83.37%

Enhanced 103,157 - 0.86%


10

Таble 1 Structure of banking sector clients by risk level and CDD actions and measures applied

Risk level No of clients Share CDD actions

and measures Number of clients Share

Low 5,782,766 48.38% Simplified 1,884,575 15.77%

Medium 6,081,324 50.87% General 9,966,214 83.37%

High 89,856 0.75% Enchanced 103,157 0.86%

11,953,946 100% 11,953,946 100.00%

Based on the analysis of the data from the Questionnaire, it can be concluded that banks do not apply simplified actions and measures to all low-risk clients, but only 32.59% of them, while the rest of such clients are subject to general CDD actions and measures. Enhanced actions and measures are applied to all clients classified as high-risk, but also to 13,301 clients classified as medium- or low-risk.

In the question from the Questionnaire about the number of cases in which the business relationship was discontinued due to the inability to apply CDD actions and measures, 13 banks responded that business relations with 36,550 clients were terminated in the period 1 January–31 March 2019. In the said period, 15 banks refused the offer to establish a business relationship and/or execute transactions in 8,247 cases, due to the impossibility to take CDD actions and measures. Of this number of cases, 8,156 pertained to one bank only.

VI Transactions and products

Banks executed a total of 143,744,614 transactions in the period 1 January–31 March 2019.

Chart 11 shows the structure of cash and non-cash (dinar and foreign currency) transactions in the banking sector in the period 1 January–31 March 2019.

The largest share in the total number of transactions executed in the entire banking sector belongs to non-cash transactions in dinars - around 75%, followed by cash transactions with the share of 23.24%, and non-cash foreign transactions with the share of 1.76%. In the same period, banks executed 836,964 non-cash and cash transactions amounting to EUR 15,000 or more in the dinar equivalent at the NBS middle exchange rate, with non-cash transactions in dinars also accounting for the largest share – almost 72%.


11

Chart 13 shows a comparative overview of the number of cash transactions worth EUR 15,000 or more in the dinar equivalent at the NBS middle exchange rate, from the aspect of whether they were performed by the clients in the bank with which they have established a business relationship or not (so-called non-clients).

Out of a total of 68,314 cash transactions worth EUR 15,000 or more in the dinar equivalent at the NBS middle exchange rate, 1,738 transactions were executed by persons who were not clients of the bank in which these transactions were made.

Chart 11 Share of cash and non‐cash transactions (in RSD and foreign currency) in the banking system in the period 1

January–31 March2019

Cash 33,412,440 - 23.24%

Non-cash in RSD 107,940,280 - 75.09%

Non-cash in foreign currency 2,391,894 - 1.66%

Chart 12 Share of cash and non-cash transactions amounting to EUR 15,000 or more in dinar equivalent in the period January–March 2019

Cash 68,314 - 81.62%

Non-cash in RSD 602,565 - 71.99%


12

Below is an overview of the number of payments from high-risk countries and offshore geographical areas executed in the period January–March 2019. In the analysed period, 37,836 payments were made to high-risk countries, as well as 7,070 payments to off-shore geographical areas, i.e. countries.

In the period January–March 2019, ten banks introduced a total of 31 new products into their business offer.

In the analysed period, 13 banks identified a high ML/TF risk for less than 10 products; 4 banks did the same for 11–30 of their products, and one bank for more than 30 products. Five banks identified a medium ML/TF risk for less

Chart 13 Composition of executed cash transactions amounting to EUR 15,000 – clients/non-clients in the period January-March 2019

Cash clients 466,576 - 97.45%

Cash non-clients 1,738 - 2.54%

0

5,000

10,000

15,000

20,000

25,000

30,000

35,000

40,000

45,000

Payments from abroad Payments to abroad

Chart 14 Number of executed payments from and payments to high-risk and offshore countries in the period January-March 2019

High risk countries


13

than 10 of their products; one bank for 11–30 of its products and two banks for more than 30 products. Six banks reported no products from their business offer were assessed as high- or medium-high ML/TF risk.

As regards electronic and mobile banking, e-banking is used by 1,895,183 clients and m-banking by 1,527,960 clients. According to the data provided, 19 out of 27 banks approved 8,339 loans worth EUR 40,951,291against 100% deposit backing.

VII Outsourcing CDD actions and measures to third parties

In the entire banking sector, 7 banks used the possibility to outsource some of the CDD actions and measures to third parties, in accordance with the Law, and thus obtained documentation for 462,861 clients (438,457 of these clients pertaining to one bank only).

VIII Correspondent relationship

In the analysed period, seven banks declared that they did not establish correspondent relationships with banks and similar institutions with a head office in a foreign country, while the remaining 20 banks established a total of 185 loro correspondent relations. Out of 27 banks, 26 declared that they did not establish loro correspondent relationship with a bank or another similar institution with a head office in a foreign country that is on the list of countries with strategic deficiencies in the AML/CFT system.

Two banks declared that they closed a total of 4 loro accounts in the period January-March 2019, citing some of the following reasons: rightsizing, no need for such an account, inability of the foreign bank to fully submit all the required documentation within the prescribed period, etc.

Banks have nostro correspondent relations with 235 banks, while three banks declared that they have established nostro correspondent relations with five banks with a head office in a foreign country that is on the list of countries with strategic deficiencies in the AML/CFT system.

IX Corporative governance

Twenty-six out of 27 banks have special software for identifying suspicious transactions and persons, which also recognises transactions in the amount of EUR 15,000 or more in the dinar equivalent, while one bank was in the process of introducing such software. Twenty-four banks use external software (Smaragd, Norcom, Asseco, SEE, Aseba AML Tool, etc.). The remaining two banks have developed AML software in-house.


14

In all 26 banks, the aforementioned software recognises several interconnected transactions amounting to a total of EUR 15,000 or more. In 22 banks, this software classifies clients according to the risk level. Asked whether the software allows filtering of clients and transactions by embargo and so-called blacklists (the UN Security Council list, the EU lists and the OFAC lists, etc.), all 27 banks responded positively.

Banks also listed the indicators included in the software which they selected from the List of indicators for recognising suspicious transactions or persons and the List of indicators relating to terrorism financing published on the Administration’s website.

Table 2 Number of included indicators relating to money laundering

Number of bank indicators Number of banks

Less than 10 7

10–19 13

20–29 6

30–37 1

The table below shows how many indicators for recognising suspicious transactions relating to money laundering published on the Administration’s website were included in the software solutions of 26 banks.

Table 3 Number of included indicators relating to terrorism financing

Number of bank indicators Number of banks

Less than 5 20

5–10 4

More than 10 2

Indicators for recognising FT-related suspicious transactions are included in the software of 26 banks, as follows:

X Employee training

In the banking sector, front-office jobs are carried out by a total of 13,249 employees. This is 157 persons less than in the previously analysed period (as at 31 December 2018), while the relative share of front-office staff in total banking sector employment is 58.33%.

Training is most often carried out by AML/CFT compliance officers and their deputies, trained managers of organisational units and Compliance Division staff. In one bank, training is carried out in direct contact with employees, two banks carry out training by E-learning, and the remaining 24


15

banks carry out trainings in both ways. A total of 19 banks stated that training is organised once a year, five banks organise these trainings twice a year, while three banks do so three or more times a year. Trainings are usually carried out as seminars, workshops, consultations and expert meetings.

As at 31 March 2019, 11,631 employees successfully completed the training. All banks stated they introduced their employees with the consequences of non-compliance with laws, procedures and findings of the ML/TF risk management control.

XI Organisational structure

Twenty-three banks have a special organisational unit which deals exclusively with the implementation of the Law, whereas in four banks these tasks are performed by other organisational units. In 23 banks in separate organisational units, a total of 103 employees have been assigned to AML/CFT-related tasks, including a compliance officer and his deputy.

When asked if the bank submitted to the Administration data on the personal name and job title of the compliance officer and his deputy, as well as the data on the personal name and job title of the member of the top management responsible for the implementation of the Law, all banks replied affirmatively.

When asked if the compliance officer and his deputy who engage in AML-related activities in accordance with the Law, perform other tasks too, 23 banks gave a negative response, while four banks declared that the compliance officer and his deputy perform other tasks as well.

In three of the 27 banks (11.11 %), the AML or front-office staff churn rate was between 10% and 50%, while in all other banks it was below 10%.

XII Reporting to the AML/CFT compliance officer

The front-office staff submitted to the compliance officer 1,833 internal reports on suspected ML risk in relation to a client or a transaction, with 1,656 of these reports pertaining to a single bank. Banks stated that the number of internal reports that bank staff submitted to the compliance officer on suspected ML risk, based on which data were not submitted to the Administration, amounted to 1,754, meaning that only 4.31% of the total number of internal reports was reported to the Administration. In the period January–March 2019, bank staff sent only one internal report to the compliance officer on suspected TF risk, which was submitted to the Administration.


16

XIII Internal audit and internal control

According to Questionnaire responses, all banks regulated the internal controls system in their internal acts. Internal control of implementation of the Law is within the remit of the organisational unit in charge of internal audit, but also of other organisational units such as unit in charge of bank supervision, unit in charge of supervising “network” operations or the unit in charge of AML/CFT activities only.

In the last year, 25 banks performed an AML/CFT-related internal audit. The remaining two banks, which failed to do so, carried out the internal audit in 2015 ad 2016. Internal audits identified deficiencies in 23 banks and set deadlines for their removal. The identified deficiencies include: AML risk assessments, incomplete documentation, incorrect certification of the copies of the personal documents of clients – natural persons and beneficial owners of legal persons, identification, inadequate client classification, inadequate application of AML tools, deficiencies related to the procedure of establishing the political exposure status of the client and the beneficial owner, reflected in untimely collection of statutory statements and implementation of a systematic solution for generating statements, as well as registering the collected data in the banking system, deficiencies in documentation needed for account opening, etc. Of the 23 banks which identified omissions in the course of internal audit, 14 removed such omissions within the set deadline, while in nine banks the deadline has not yet expired.

In the last year, 26 banks performed an AML/CFT-related internal control. One bank, which failed to do so, carried out the last such internal control in Q3 and Q4 2017. Irregularities were detected in 25 banks in the course of internal control and deadlines were set for the removal of such irregularities. The most frequent irregularities related to the following: technical omissions, incomplete client documentation, inadequate client classification, inadequate identification of clients and beneficial owners, non-complying AML risk assessment, deficiencies regarding the origin of property, etc. Of the 25 banks which identified irregularities in the course of internal control, 22 removed such irregularities within the set deadline, while in three banks the deadline has not yet expired.

In the analysed period, four banks stated they are conducting a total of five disciplinary procedures due to problems with employee integrity (participation in fraud, theft, corruption, etc.), while three banks stated that procedures are conducted against the bank or employees in respect of ML/TF.

Documents

ANALYSIS OF BANKS’ RESPONSES TO THE QUESTIONNAIRE ON ... · Questionnaire, regarding the implementation of measures for the ML/TF risk management and control in the period from