Payment flows of interbank payment and settlement systems in 2009

Payments and Securities Settlements Payment flows of interbank payment and settlement systems in 2009 Prepared by: Anna Morvay (MNB 1563) The annual report about the payment flows in interbank systems analyses the year 2009 turnover of the systems operated at the MNB and of the ICS operated by GIRO in the following framework: I. Data and charts relative to all payment transactions (VIBER, ICS, InForex) II. Data and payment flows of individual payment systems 1. Payment flows in the ICS 2. Payment flows in VIBER 3. Payment flows in VIBER and the MNB s home accounting system 4. Payment transactions carried out by the Hungarian Post 1 III. Comparison of the payment flows of individual systems The charts and tables mainly contain time series in annual, quarterly, monthly or sometimes daily breakdowns. In addition to the time series, important parts are the distribution of the turnover from several aspects, the ratio of individual payment types to one another and the intraday distribution of the turnover. This paper discusses liquidity issues, queuing and its intraday pattern. At several points it emphasises the data typical of the five largest banks compared with all the data. I. Data and charts relative to all payment transactions (VIBER, ICS, InForex) Interbank payment transactions are carried out in the ICS operated by GIRO Zrt. since 1994 and in the systems of the MNB. Only a home accounting system had been operated in the Central Bank earlier. Realtime settlement of largevalue payments has been possible since 1999 in VIBER, which currently settles payments together with the home accounting system (InForex). The task of the ICS is to generate the clearing position of the payment transactions sent to GIRO, settlement is done at the MNB. Chart 1 shows that the total turnover of the three systems together has grown 7.5 times higher in terms of volume and 36 times higher in terms of value since 1995. Initially, the increase in terms of volume was higher, then the value of the turnover grew more and more rapidly, in which foreign bank clients appearing in the turnover of VIBER played a significant role. The effect of the crisis is seen in the payment turnover as well: from 2008 to 2009 the increase in the number of payments considering all the systems was minimal (1%), while the value of the turnover fell by nearly 4%, breaking the earlier increasing trend. 1 The Post is not considered to be a payment system.

Chart 1: Value and volume of interbank payment turnover (ICS + VIBER + InForex) 1 200 1 000 984 220 million peaces 300 1 119 1 076 240 242,5 250 205 879 800 188 200 600 126 140 155 162 173 515 650 150 103 453 400 200 306 63 50 41 185 32 125 55 71 85 30 37 1995. 1996. 1997. 1998. 1999. 2000. 2001. 2002. 2003. 2004. 2005. 2006. 2007 2008 2009 100 50 HUF billions million peaces The payment turnover in the year under review was 42 times of the GDP estimated for 2009 (earlier data are compared to actual GDP). It can be observed (Chart 2) that there is a slowdown in the increase in this ratio; the minimum growth in 2009 was a result of the fact that compared to the previous year the decline in the estimated GDP was greater than in the payment turnover. Chart 2: The payment turnover to GDP ratio by all systems (ICS, VIBER, InForex) Interbank turnover/gdp 45 42,1 42,5 40 35 37,0 38,7 30 29,6 25 24,1 24,8 20 17,8 15 12,1 10 5 5,2 6,3 6,8 7,3 9,4 1996. 1997. 1998. 1999. 2000. 2001. 2002. 2003. 2004. 2005. 2006. 2007. 2008. 2009. Examining the growth as a proportion of GDP for individual payment systems it can be established that the changes in the ratio are mainly influenced by the decline in the turnover of

VIBER, while significant turnover became rechanneled into the MNB s home accounting system (mainly because of the preference for transactions concluded with the Central Bank). Chart 3: Payment transactions as a proportion of GDP by systems Interbank turnover/gdp 1,4 2,9 40,0 35,0 1,5 0,3 30,0 1,2 25,0 0,4 0,7 20,0 0,1 32,9 35,8 38,1 37,2 15,0 0,2 21,1 21,6 25,9 10,0 15,0 9,1 5,0 2,8 2,7 2,6 2,5 2,5 2,6 2,6 2,6 2,5 2001 2002 2 003 2004 2005 2006 2007 2 008 2009 ICS VIBER Home accounting system In terms of volume, the overwhelming majority of transactions is settled in the ICS, typically as a result of the high number of lowvalue customer payments sent to GIRO. However, in terms of value, most of the turnover is settled in VIBER. Table 1: Changes in payment transactions by systems in terms of volume and value 2001. 2002. 2003. 2004. 2005. 2006. 2007. 2008. 2009. Number of payments thousend ICS 139 573 154 482 161 855 172 335 186 732 204 069 219 375 238 586 241 225 VIBER 240 337 430 555 676 809 892 1 241 981 NBH home accou 50 58 71 88 125 178 195 236 281 Total 139 863 154 877 162 356 172 978 187 533 205 056 220 462 240 064 242 487 Value of payments ICS 42 985 45 074 48 456 51 320 55 610 61 887 65 907 70 335 63 482 VIBER 139 757 258 931 396 633 449 798 568 652 781 645 910 637 1 011 281 939 929 NBH home accou 2 588 1 782 7 622 13 778 26 147 35 339 7 604 36 980 72 258 Total 185 331 305 788 452 711 514 896 650 409 878 871 984 149 1 118 596 1 075 669 Examining the value of the turnover in 2008 and 2009 in a monthly breakdown compared to the annual average reveals that the seasonal fluctuation altered; the aboveaverage turnover in March and April is followed by a declining trend, which is well below the average according to Chart 4.

2008 január 2009 január 2008 január 2009 január Chart 4: Total (VIBER + InForex + ICS) monthly interbank turnover in terms of value compared to the annualised monthly average 120 000 100 000 80 000 60 000 40 000 20 000 Interbank turnover Average turnover In terms of volume, the fluctuation in 2009 is lower than the annual average, and the turnover in the last months of the year increased more slowly than in the previous year. Chart 5: Total (VIBER + InForex + ICS) monthly average interbank turnover in terms of volume compared to the annualised monthly average thousend peaces 1 200,0 1 000,0 800,0 600,0 400,0 200,0 Interbank daily turnover Daily everage

Chart 6 shows the average amount of transactions (in HUF million for the MNB s systems and HUF thousand for the ICS). In the case of VIBER, the earlier increasing trend has already been followed by a decline for two years (the average amount of transactions fell to below HUF 1 billion). A very slight steady decline has been typical in the ICS since 1999. The increase in the average amount of transactions of the home accounting system was the result of the foreign exchange and deposit transactions with the Central Bank. Chart 6: Average amount of transactions in individual payment systems HUF million, HUF thousend, HUF million 1 200,0 1 000,0 800,0 600,0 400,0 200,0 495,2 449,5 513,8 536,2 362,0 503,9 517,2 582,7 767,6 922,0 810,5 840,9 965,7 333,2 307,7 291,8 299,0 298,0 297,8 303,3 32,3 51,8 30,7 106,8 156,4 209,7 1 020,5 300,4 1 006,0 958,4 294,8 263,2 257,1 198,7 39,1 156,5 1995. 1996. 1997. 1998. 1999. 2000. 2001. 2002. 2003. 2004. 2005. 2006. 2007. 2008 2009 VIBER HUF mns/transaction Home accounting HUF mns/transaction ICS HUF thousend/transaction According to Chart 7, in the year following the launching of VIBER the annual average of the total daily liquidity of VIBER members had exceeded the payment turnover to be carried out; then the trend turned, and the turnover kept on increasing compared to the available liquidity. The trend changed in 2009: the still increasing liquidity was coupled with a decline in payment transactions. The payment turnover to liquidity ratio was 3.48 in 2008; this figure changed to 2.71 in 2009. The share of account balance within VIBER members liquidity declined (as a consequence of the decline in the level of the reserve requirement), while total credit line is increasing continuously. It means that banks were in a more comfortable liquidity situation as a result of the higher liquidity compared to the turnover.

2004. január 2005. január 2006. január 2007. január 2008. január 2009. január Chart 7: Annual changes in average daily interbank payment transactions (ICS + VIBER + InForex) and liquidity (money on account + credit line) between 2000 and 2009 4500 4407 4268 4000 3500 3490 3924 3000 2500 2000 1500 1000 500 0 2551 2026 1812 319 665 1220 591 429 475 537 534 393 739 384 378 501 1257 45 86 201 303 412 372 395 538 648 2000 2000 2001 2001 2002 2002 2003 2003 2004 2004 2005 2005 2006 2006 2007 2007 2008 2008 2009 2009 daily turnover account balance credit line Looking at the monthly data, there is a well visible sudden increase taking place in November 2008 in credit line at the expense of account balance, which is declining continuously in 2009, while payment turnover is clearly falling. Chart 8: Monthly changes in average daily interbank payment transactions (ICS + VIBER + InForex) and liquidity (account balance + credit line) between 2004 and 2009 5 000 4 000 3 000 2 000 1 000 daily turnover account balance credit line

2000. január 2001. január 2002. január 2003. január 2004. január 2005. január 2006. január 2007. január 2008. január 2009. január The same phenomenon is depicted in the chart following the payment turnover to liquidity ratio, where the decline for 2009 is spectacular. Chart 9: Monthly changes in the ratio of average daily interbank payment transactions (ICS + VIBER + InForex) to liquidity (account balance + credit line) between 2003 and 2009 5,00 4,50 4,00 3,50 3,00 2,50 2,00 1,50 1,00 0,50

1995 január 1996 január 1997 január 1998 január 1999 január 2000 január 2001 január 2002 január 2003 január 2004 január 2005 január 2006 január 2007 január 2008 január 2009 január II. Data and payment flows of individual payment systems 1. Payment flows in the ICS The annual turnover graph of the ICS follows the general trend: by 2009 the continuous growth in terms of volume decelerated suddenly, while a fall in value was recorded. Chart 10: Volume and value of payments in the ICS HUF mns 80 000 thousend peaces 300 000 70 000 250 000 60 000 50 000 200 000 40 000 150 000 30 000 100 000 20 000 50 000 1995. 1996. 1997. 1998. 1999. 2000. 2001. 2002. 2003. 2004. 2005. 2006. 2007. 2008. 2009. ICS turnover value volume thousend peaces The annual change is reflected in the monthly data, although an express increase in both monthly value and volume is experienced at end2009. Chart 11: Changes in the monthly value and volume of payment orders settled in the ICS 8 000 thousend peaces 25 000 7 000 6 000 20 000 5 000 15 000 4 000 3 000 2 000 5 000 1 000 Value Volume

2003. január 2004. január 2005. január 2006. január 2007. január 2008. január 2009. január The monthly average of turnover per day also follows the above pattern: the fluctuations in volume show a constant but increasing trend, the increasing trend of the turnover ceased to exist by 2009, and decline is typical until December. Chart 12: Daily average of the value and volume of payments settled in the ICS every month in 2003 2009 400 350 300 250 thousend peacs 1 200 1 000 800 200 600 150 100 50 400 200 Value of turnover Volume of turnover The distribution of the ICS payment turnover by value ranges reflects the intended purpose of the system: it is for settling high volumes of lowamount payments. More than half of the turnover is composed of items with an individual value of less than HUF 50 million, although there are many items with high individual value (e.g. above HUF 5 billion), which would belong to the category of the realtime system. Chart 13: Payment transactions in the ICS according to individual value in 2009 HUF 10005000 mns 9% above HUF 5000 mns 8% HUF 01 mns 20% HUF 5001000 mns 4% HUF 100500 mns 12% HUF 15 mns 17% HUF 50100 mns 6% HUF 1050 mns 16% HUF 510 mns 8%

In terms of volume, the smallest items (below HUF 1 million) are undoubtedly predominant; as the value increases, the number of items decreases. Table 2: Distribution of payments settled in the ICS according to value ranges Value ranges Volume of ICS turnover in numbers 2003. dec. 2004. dec. 2005. dec. 2006. dec. 2007. dec. 2008. dec. 2009. dec HUF 01 millions 9 192 223 9 324 206 9 667 705 10 191 445 10 964 395 12 437 457 12 684 580 HUF 15 millions 447 876 436 547 454 213 466 339 486 470 516 296 467 879 HUF 510 millions 60 534 60 794 65 617 68 363 73 304 76 877 66 791 HUF 10100 millions 46 064 46 004 51 508 52 178 55 655 57 341 51 128 HUF 100500 millions 3 174 3 086 3 497 3 713 4 035 4 027 3 633 HUF 5001000 millions 355 352 331 380 422 453 353 HUF 10005000 million 210 214 243 263 298 305 245 over 5000 millions 28 35 52 45 42 41 15 total 9 750 464 9 871 238 10 243 166 10 782 726 11 584 621 13 092 797 13 274 624 Orders that remain queuing after the night processing or orders transmitted late are processed by GIRO in a second round in the morning if sufficient collateral has been provided. In the last two years it happened 15 and 16 times, respectively; the value of the items that remained in the queue tripled (increasing from HUF 32.2 billion to HUF 98.9 billion). The time series of morning processing is fluctuating both in terms of volume and value. The outlier in 2006 is the result of a largeamount late transmission. Chart 14: Annual volume and value of processing in the morning in the ICS HUF mns 600 000 574 207 number 35 500 000 27 30 26 27 30 400 000 21 23 25 20 300 000 15 259 957 14 15 16 15 200 000 100 000 154 379 141 568 98 940 77 175 64 593 32 258 13 360 14 623 2000. 2001. 2002. 2003. 2004. 2005. 2006. 2007. 2008 2009. ICS second cycle turnover value Number of ICS second cycle 10 5 In the ICS, the value of rejected payments compared to the total turnover shows a declining trend, which broke in 2007 2008.

1995. január 1996. január 1997. január 1998. január 1999. január 2000. január 2001. január 2002. január 2003. január 2004. január 2005. január 2006. január 2007. január 2008. január 2009. január Chart 15: The ratio of items rejected in the ICS according to value % 1,00 0,90 0,80 0,70 0,60 0,50 0,40 0,30 0,20 0,10

2000. I. n.év 2001. I. n.év 2002. I. n.év 2003. I. n.év 2004. I. n.év 2005. I. n.év 2006. I. n.év 2007. I. n.év 2008. I. n.év IV. n. év 2009. I. n.év Payment flows in VIBER There was a significant change in VIBER membership in 2009: as a result of the implementation of the MNB s account keeping strategy, the number of direct members increased to 57; at the same time, indirect membership through the MNB ceased to exist. This had an insignificant effect on the annual turnover as the changes to direct membership took place in the second half of the year, and new members earlier may also have transmitted VIBER orders through the MNB as indirect participants. Looking at the time series of the annual turnover of VIBER, the steady growth that had started upon launching came to a halt as a result of the crisis, and for the first time a fall was observed both in terms of value and volume by 2009. Chart 16: Annual volume and value of payment transactions settled in VIBER peaces 1 200 000 1 000 000 892 387 910 637 809 443 1 011 281 1 005 233 939 929 980 752 1 200 000 1 000 000 800 000 676 218 781 645 800 000 600 000 400 000 554 950 449 798 429 999 337 340 396 633 258 931 568 652 600 000 400 000 239 830 200 000 158 783 82 130 139 758 200 000 2000. 2001. 2002. 2003. 2004. 2005. 2006. 2007. 2008 2009. value volume peaces With regard to the quarterly time series, the effect of the crisis resulted in a more declining trend from mid2008 on. Chart 17: Quarterly value and volume of payment transactions settled in VIBER thousend peaces 300 HUF thousend bns 300 250 250 200 200 150 150 100 100 50 50 value HUF thousend bns volume thousend peaces

2009/01/05 2009/01/12 2009/01/19 2009/01/26 2009/02/02 2009/02/09 2009/02/16 2009/02/23 2009/03/02 2009/03/09 2009/03/16 2009/03/23 2009/03/28 2009/04/03 2009/04/10 2009/04/20 2009/04/27 2009/05/05 2009/05/12 2009/05/19 2009/05/26 2009/06/03 2009/06/10 2009/06/17 2009/06/24 2009/07/01 2009/07/08 2009/07/15 2009/07/22 2009/07/29 2009/08/05 2009/08/12 2009/08/19 2009/08/28 2009/09/03 2009/09/10 2009/09/17 2009/09/24 2009/10/01 2009/10/08 2009/10/15 2009/10/22 2009/10/30 2009/11/06 2009/11/13 2009/11/20 2009/11/27 2009/12/04 2009/12/11 2009/12/18 2009/12/28 2000. január 2001. január 2002. január 2003. január 2004. január 2005. január 2006. január 2007. január 2008. január 2009. január Starting from 2007, the monthly turnover of VIBER shows increasingly strong fluctuations; the declining trend is observed in terms of both value and volume in 2009. Chart 18: Monthly value and volume of payment transactions settled in VIBER HUF thousend bns 120 thosend peaces 120 100 100 80 80 60 60 40 40 20 20 volue HUF thousend bns volum thousend peaces Chart 19 shows the daily number of payment transactions settled in VIBER compared to the monthly average. The outstanding lows in turnover indicate the low turnover of Saturdays that were working days. Chart 19: The daily number of payments settled in VIBER compared to the monthly average in 2009 daily volume peaces 6000 5000 4000 3000 2000 1000 0 daily volume monthly average

2004. 2005. 2006. 2007. 2008. 2009 január 2009.01.05 2009.01.09 2009.01.15 2009.01.21 2009.01.27 2009.02.02 2009.02.06 2009.02.12 2009.02.18 2009.02.24 2009.03.02 2009.03.06 2009.03.12 2009.03.18 2009.03.24 2009.03.28 2009.04.02 2009.04.08 2009.04.15 2009.04.21 2009.04.27 2009.05.04 2009.05.08 2009.05.14 2009.05.20 2009.05.26 2009.06.02 2009.06.08 2009.06.12 2009.06.18 2009.06.24 2009.06.30 2009/07/06 2009/07/10 2009/07/16 2009/07/22 2009/07/28 2009.08.03 2009.08.07 2009.08.13 2009.08.19 2009.08.27 2009/09/01 2009/09/07 2009/09/11 2009/09/17 2009/09/23 2009/09/29 2009/10/05 2009/10/09 2009/10/15 2009/10/21 2009/10/28 2009/11/03 2009/11/09 2009/11/13 2009/11/19 2009/11/25 2009/12/01 2009/12/07 2009/12/11 2009/12/17 2009/12/22 An analysis of the value of the daily turnover reveals that there are more outliers above the average than below it. Without exception, the first fifty highest daily turnover values fall on a Wednesday. Chart 20: The daily value of payments settled in VIBER compared to the monthly average in 2009 /day 8 000 7 000 6 000 5 000 4 000 3 000 2 000 1 000 daily turnover daily average turnover Chart 21 shows the monthly turnover of banks in an order of magnitude according to their average monthly turnover. The annual turnovers of 19 banks exceed 10 thousand payments per year; other banks have a lower turnover. The monthly developments in VIBER turnover are determined by the pattern of the five banks with the highest turnover. Chart 21: Changes in the turnover of the five largest banks compared to the value of all the payments settled in VIBER 100 000 90 000 80 000 70 000 60 000 50 000 40 000 30 000 20 000 value of total turnover in VIBER turnover of the five largest banks

2002.I.30. II.20. III.27. IV.17. V.14. VI.12. VII.10. VII.14. IX.18. X.24. XI.25. XII.27. 2003.I.30. II.20. III.27. IV.17. V.14. VI.12. VII.10. VIII.14. IX.18. X.24. XI.25. XII.27. 2004.I.21 II.17 III.17 IV.07 V.05 VI.09 VII.21 VIII.24 IX.30 X.20 XI.24 XII.22 2005.I.18 II.22. III.16. IV.13. V.25. VI.15. VII.13 VIII.24 IX.29 X.19 XI.02 XII.14 2006.I.31 II.24. III.16. IV.13. V.30. VI.21. VII.05 VIII.24 IX.20 X.24. XI.24 XII.20 2007.I.24 II.28. III.21. IV.25. V.02. VI.20. VII.25 VIII.22 IX.28 X.31. XI.21 XII.19 2008.I.23 II.27. III.19 IV.09 V.14. VI.18 VII.30 VIII.13 IX.17 X.14 XI.12 XII.17 2009.I.07 II.25 III.18 IV.22 V.20 VI.24 VII.08 VIII.26 IX.30 X.07 XI.18 XII.16 2002.01.31 02.28. 03.20. 04.17. 05.2. 06.12. 07.10. 08.2. 09.18. 10.24. 11.21 12.27. 2003.01.21 02.27. 03.13. 04.4. 05.14. 06.6. 07.7. 08.21. 09.9. 10.15. 11.12 12.4. 2004.01.21 02.21. 03.17 04.05 5. 3. 6. 9. 7. 23. 8. 24. 9. 15. 10. 18. 11. 24. 12. 15. 2005.01.12 02.21. 03.16. 04.08. 05.31. 06.15. 07.5 08.24 09.09 10.19 11.14 12.12 2006.01.11 02.24. 03.16. 04.05. 05.05. 06.30. 07.5 08.31 09.20 10.10 11.24 12.20 2007.01.16 02.28. 03.30. 04.27. 05.02. 06.20. 07.31 08.21 09.28 10.24. 11.30 12.19 2008.01.23 02.28. 03.19 04.30 05.09 06.18 07.17 08.11 09.17 10.14 11.12 12.17 2009.01.12 02.27. 03.18 04.14 05.11 06. 17 07.29 08.08 09.16 10.13 11.04 12.16 Chart 22 depicts the time series of the monthly peak days in terms of volume. To date, the largest volume of payments on a single day was recorded on 14 October 2008 with 8 054 transactions. Chart 22: Number of transactions settled in VIBER on monthly peak days number of transactions peaces 8 000 7 000 6 000 5 000 4 000 3 000 2 000 1 000 In terms of value, VIBER reached the record of monthly peak days on 7 October 2009 with a daily turnover of HUF 10 174 billion. Chart 23: Value of transactions processed in VIBER on monthly peak days 8 000 6 000 4 000 2 000

The distribution of the number of VIBER transactions according to value ranges shows that more than 60% of orders have an individual value of less than HUF 100 million, and only 4% of the transactions falls in the range of above HUF 5 billion. Chart 24: Breakdown of the volume of VIBER transactions according to value ranges in 2009 above HUF 5000 mns 4% HUF 10005000 mns 7% below HUF 10 mns 41% HUF 1001000 mns 28% HUF 10100 mns 20% Transactions with an individual value exceeding HUF 5 billion, which only represent 4 % in terms of volume, account for 74% of the value of all VIBER transactions. Chart 25: Breakdown of the value of payments in VIBER according to value ranges in 2009 below HUF 10 mns 0% HUF 10100 mns 6% HUF 1001000 mns 10% above HUF 5000 mns 74% HUF 10005000 mns 10%

Examining the monthly data of the last two years according to the types of payment orders, banktobank items have the largest share. From the autumn of 2008 it declined continuously (except for the peak in March), while the share of transactions concluded with the Central Bank increased. Compared to the level of 2008 the value of securities transactions on the DVP principle also declined; although the value of the turnover increased by the second half of 2009, it did not reach the level of the previous year. Chart 26: Breakdown of payments settled in VIBER according to types of transactions, in value HUF thousend bns 90,0 80,0 70,0 60,0 50,0 40,0 30,0 20,0 10,0 customer payments transactions with the Central Bank DVP securities transactions banktobank items The composition of payment orders by the number of transactions shows a different trend compared to the one by value. As regards the respective shares, banktobank transactions are followed by customer payments; securities transactions have a smaller share, while the number of transactions concluded with the Central Bank is the lowest. In the two years under review, all payment orders were at their peaks in October 2008, followed by a significant declining trend. By end2009 only the number of customer payments exceeded the peak reached in October 2008, while the number of other types of transactions was lower.

2008 január 2009 január Chart 27: Breakdown of payments settled in VIBER according to types of transactions, in volume thousend peaces 70 60 50 40 30 20 10 0 transactions with the Central Bank customer payments DVP securities transactions banktobank items In terms of value, the annual data of securities transactions (excluding MNB bonds) settled in line with the DVP principle show a sharp increase until 2007, followed by a significant (43%) fall by 2009. In terms of volume, the growth lasted until 2008; the decline in 2009 was moderate (3.8%). Chart 28: DVP transactions in VIBER, 2000 2009 peaces 200 000 180 000 160 000 140 000 120 000 100 000 80 000 60 000 40 000 20 000 46 006 14 627 59 135 19 895 79 298 93 537 23 883 114 082 32 166 139 113 40 015 165 165 173 330 157 375 166 758 56 678 129 825 81 056 111 030 140 000 120 000 100 000 80 000 73 337 60 000 40 000 20 000 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 volume value The examination of the monthly changes in deposit and bond transactions with the MNB reveals that the value of transactions concluded with the MNB before 2007 is not significant. Following

2004. január 2005. január 2006. január 2007. január 2008. január 2009. január the introduction of the MNB bond in 2007 the value of transactions grew steadily; a sudden change is observed after September 2008. Depositing prevailed at this time, with a subsequent gradual shift to bonds. The total turnover was in the range of HUF 20 25 thousand billion in almost each month from December 2008 on. 25 000 Chart 29: Deposit and bond turnover with the MNB in VIBER 20 000 15 000 5 000 Deposit with MNB MNB bonds Payment transactions initiated by banks customers in VIBER show an increase both in terms of value and volume, although the rate of growth decelerated in recent years. Chart 30: Annual changes in customer transfers in VIBER 35 000 30 000 25 000 20 000 13 700 13 233 19 779 126 536 24 570 139 985 28 550 160 425 30 371 31 151 215 418 257 045 peaces 300 000 250 000 200 000 150 000 15 000 35 242 9 861 10 445 42 921 65 128 93 634 100 000 5 000 1 702 8 845 2000 2001 2002 2003 2004 2005 2 006 2007. 2 008 2 009 50 000 value volume

2000. 2001. január 2002. január 2003. január 2004. január 2005. január 2006. január 2007. január 2008. január 2009. január The monthly figures of the number and value of customer payments show a fluctuating but increasing trend. Chart 31: Monthly changes in customer transfers in VIBER 30 000 4000 25 000 3500 3000 20 000 2500 15 000 2000 1500 1000 5 000 500 0 value volume The concentration ratio means the size of the share of the turnover of the five largest VIBER participants compared to all members in terms of volume and value separately. Based on the number of transactions, the concentration of payments in VIBER increased until 2007; it has been significantly declining since then year by year. In terms of value, the growth lasted until 2006; the share of the five largest banks fell to 54.2% since then. Table 4: Concentration of payment orders in VIBER volume value 2003. 2004. 2005. 2006. 2007. 2008. 2009. 2003. 2004. 2005. 2006. 2007. 2008. 2009. 50,92 52,42 52,57 55,33 57,93 56,41 53,06 57,78 61,72 63,88 66,16 63,26 64,16 54,21 Chart 37 shows the intraday hourly distribution of payments settled in VIBER in value and volume, calculated from December 2009 data including the MNB. The intraday distribution follows a similar pattern: the higher turnover in the morning is a result of payment orders given with a value date and the settlement of overdue central bank transactions. Turnover picks up after 9 o clock, while the peak hour is between 10 and 11; the turnover declines steadily after that. Very few transactions are performed in the last operating hour.

Chart 32: Intraday distribution of payments processed in VIBER in December 2009 16 000 peaces 14 000 14 000 12 000 12 000 8 000 8 000 6 000 6 000 4 000 4 000 2 000 2 000 7 8 óra 8 9 óra 9 10 óra 10 11 óra 11 12 óra 12 13 óra 13 14 óra 14 15 óra 15 16 óra 16 17 óra VIBER monthly turnover value VIBER monthly turnover volume Both the value and volume of payment orders queuing up in VIBER 2 increased considerably before starting to decline gradually from 2006 on. Chart 33: The value of queuing amounts and the number of queuing in VIBER, 2000 2009 100 000 90 000 80 000 70 000 60 000 50 000 40 000 30 000 1137 2227 2717 2807 32 266 29 173 2637 33 230 71 904 3599 89 959 3856 61 066 2781 number of queuing 4500 4000 3500 3000 48 308 2500 1871 2000 31 227 1063 1500 20 000 11 997 1 253 2000. év 2001. év 2002. év 2003. év 2004. év 2005. év 2006. év 2007.év 2008.év 2009. év Value Number of queuing 1000 500 0 2 Total of the amounts in the messages sent about the items at the front of the queue. In the event that the bank already has a queue, a message is sent about the further items behind it in the queue only if the item at the front changes after 90 seconds (i.e. the number and value of the items queuing up is greater than that).

2000. január 2001. január 2002. január 2003. január 2004. január 2005. január 2006. január 2007. január 2008. január 2009. január 2000. I. n.év 2001. I. n.év 2002. I. n.év 2003. I. n.év 2004. I. n.év 2005. I. n.év 2006. I. n.év 2007. I. n.év 2008. I. n.év 2009. I. n.év As shown in Chart 34, the declining trend revealed quarter by quarter turned into a slight increase in 2009. Chart 34: Quarterly volume and value of transactions queuing in VIBER 30 000 number 1 200 25 000 1 000 20 000 800 15 000 600 400 5 000 200 Value of queuing transactions Number of opening queues Monthly data on queuing show that until mid2005 payment orders had queued up in a larger volume and lower total value, then the value of the queues increased in several waves. In terms of value, the highest queuing was recorded in November 2006 (HUF 10.6 thousand billion), while in terms of volume the peak was reached in June 2006 (queuing at banks evolved on 374 occasions). 2007 was characterised by continuous fluctuations; in 2008 a clear declining trend was observed until November, followed by an increasing trend again. Accordingly, the effect of the liquidity crisis did not result in an increase in queuing measured at a monthly level during the whole period of the crisis (no information is available on the unsubmitted orders queuing in banks drawers ). Chart 35: Monthly volume and value of transactions queuing in VIBER number of queuing 400 350 8 000 300 250 6 000 200 4 000 150 100 2 000 50 Value of queuing transactions Number of opening queues

2000. I. n.év 2001. I. n.év 2002. I. n.év 2003. I. n.év 2004. I. n.év 2005. I. n.év 2006. I. n.év 2007. I. n.év 2008. I. n.év 2009. I. n.év The comparison of the VIBER turnover and the sum of the queues shows that initially the value of the queues was below 10% of the value of the turnover, reaching the highest value in 2005 Q2 (16.3%). From 2008 on this ratio declined to below 5%, which indicates an improving liquidity situation (there is more liquidity available for conducting payment transactions). Chart 36: Value of transactions queuing in VIBER compared to the turnover by quarters 300000 250000 200000 150000 100000 50000 0 Value of queuing Total value of turnover Chart 37 shows that the developments in the turnover of VIBER and the pattern of queuing are mainly determined by the five banks with the highest turnover. Chart 37: The share of the five banks with the highest turnover in the turnover of VIBER and in the queuing amounts 100 000 90 000 80 000 70 000 60 000 50 000 40 000 30 000 20 000 value of total turnover in VIBER Value of total queu in VIBER turnover of the five largest banks Value of queu of five largest banks

2009/01/05 2009/01/08 2009/01/13 2009/01/16 2009/01/22 2009/01/27 2009/01/30 2009/02/04 2009/02/09 2009/02/12 2009/02/18 2009/02/23 2009/02/26 2009/03/03 2009/03/06 2009/03/11 2009/03/16 2009/03/19 2009/03/24 2009/03/27 2009/04/01 2009/04/06 2009/04/09 2009/04/16 2009/04/21 2009/04/24 2009/04/29 2009/05/05 2009/05/08 2009/05/13 2009/05/18 2009/05/21 2009/05/26 2009/05/29 2009/06/04 2009/06/09 2009/06/12 2009/06/17 2009/06/22 2009/06/25 2009/06/30 2009/07/03 2009/07/08 2009/07/13 2009/07/16 2009/07/21 2009/07/27 2009/07/30 2009/08/04 2009/08/07 2009/08/12 2009/08/17 2009/08/24 2009/08/27 2009/09/01 2009/09/04 2009/09/09 2009/09/14 2009/09/17 2009/09/22 2009/09/25 2009/09/30 2009/10/05 2009/10/08 2009/10/13 2009/10/20 2009/10/26 2009/10/29 2009/11/03 2009/11/06 2009/11/11 2009/11/16 2009/11/19 2009/11/24 2009/11/27 2009/12/02 2009/12/07 2009/12/10 2009/12/15 2009/12/18 2009/12/22 2009/12/29 Payment orders are settled in an average 56.2 seconds in VIBER, and 98% of all transactions are settled within five minutes. 3 In the case of payment orders settled in more than five minutes we presume that they queue up for lack of funds. In 2009 the highest number of queuing transactions (305 ea) was recorded on 17 June with a total value of HUF 511 billion, while the highest amount (HUF 1 158 billion) was lining up on 26 May (222 transactions). The pattern of daily queues by value and volume is shown in Chart 38. Chart 38: Daily volume and value of transactions settled in VIBER in more than five minutes 1 200 number of transactions 300 1 000 250 800 200 600 150 400 100 200 50 value of daily queuing number of queuing transactions The intraday distribution of queuing is shown in Charts 39 and 40. Regarding the value of queues, the beginning of the business day starts smoothly ; the critical period is between 10 and 11 a.m. (this same period is the rush hour period of the five banks with the highest turnover). Queuing drastically falls in the next hour, then declines at an even pace towards the end of the day. 3 Calculated from the June 2009 data.

Chart 39: Intraday distribution of VIBER transactions in value settled in more than five minutes in 2009 (annual data) 14 907 14 000 12 000 8 000 7 562 6 351 6 000 4 000 2 727 2 000 877 2 080 1 643 1 140 13 58 5 78 óra 8 9 óra 9 10 óra 10 11 óra 11 12 óra 12 13 óra 13 14 óra 14 15 óra 15 16 óra 16 17 óra Extended opening hours Chart 40: Intraday distribution of VIBER transactions in volume settled in more than five minutes in 2009 (annual data) number of transactions 5 000 4 500 4 585 4 723 4 000 3 500 3 000 2 500 2 240 2 000 1 798 1 500 1 495 1 000 500 27 273 484 144 22 8 78 óra 8 9 óra 9 10 óra 10 11 óra 11 12 óra 12 13 óra 13 14 óra 14 15 óra 15 16 óra 16 17 óra Extended opening hours The value and number of transactions rejected because of various errors in the payment orders and not owing to endofday lack of funds continues to show strong fluctuations; it did not attain a declining trend especially in terms of volume.

2004. Január Február Március Április Május Június Július Augusztus Szeptember Október November December 2005. Január Február Március Április Május Június Július Augusztus Szeptember Október November December 2006. Január Február Március Április Május Június Július Augusztus Szeptember Október November December 2007. Január Február Március Április Május Június Július Augusztus Szeptember Október November December 2008. Január Február Március Április Május Június Július Augusztus Szeptember Október November December 2009. Január Február Március Április Május Június Július Augusztus Szeptember Október November December Chart 41: Value and number of payment orders rejected in VIBER 300,0 300 250,0 250 200,0 200 150,0 150 100,0 100 50,0 50 0 Value of rejected transactions Number of rejected transactions peaces The main reason for rejection is the incorrect identification of the sending/receiving bank (wrong BIC); payments were submitted twice in nearly 20% of the cases, and the wrong setting of operating hours and value date also happens frequently. Chart 42: Breakdown of transactions in value rejected in VIBER on the basis of the reason for rejection in 2009 Settlement processing finished; 5% Duplicate payment.; 18% Entry ended.; 8% Invalid value date; 0% Past payment value date.; 6% Payment to or from an invalid member.; 63% In terms of volume, the wrong value date and overdue cutoff time as well as double submissions cause most of the problems.

2007 Január Február Március Április Május Június Július Augusztus Szeptember Október November December 2008 Január Február Március Április Május Június Július Augusztus Szeptember Október November December 2009 Január Február Március Április Május Június Július Augusztus Szeptember Október November December Chart 43: Breakdown of transactions by volume rejected in VIBER on the basis of the reason for rejection in 2009 Payment to or from an invalid member; 22% Settlement processing finished; 1% Duplicate payment; 18% Entry ended; 19% Past payment value date; 39% Invalid value date; 1% The availability of VIBER (the ratio of operation to total operating hours) shows an improving trend. In 2009 the ratio was continuously above the expected 99.7%. Chart 44: VIBER availability % 100,00 99,80 99,7 99,60 99,97 99,76 99,74 99,77 99,67 99,85 99,66 99,87 99,64 99,97 99,96 99,96 99,91 99,94 99,40 99,52 99,40 99,20 99,00 99,18 99,29 98,80 98,60 98,65 98,76 98,40 98,20 98,26

2008 január 2009 január 2. Payment flows in VIBER and the MNB s home accounting system As it is depicted in Chart 45, the number of transactions settled in InForex has been increasing steadily since 2001. In terms of value, in 2007 a sudden fall was caused by the changing of the twoweek deposit into central bank bond, the settlement of which is done in VIBER. As a result of the crisis, the turnover surged again in 2008 and 2009; banks concluded transactions with the Central Bank instead of banktobank transactions. Chart 45: Annual volume and value of transaction performed in InForex 80 000 thousend peaces 300 70 000 250 60 000 50 000 200 40 000 150 30 000 100 20 000 50 2001. 2002. 2003. 2004. 2005. 2006. 2007. 2008. 2009. Value Volume Chart 46 shows the value of transactions settled in InForex in a monthly breakdown. A sharp increase in turnover starting from October 2008 is well visible (while there was a decline in the other systems /VIBER, ICS, Hungarian Post/). Since mid2009 a declining trend has been observed here as well. The change was driven by the deposit and foreign exchange transactions concluded with the MNB. Chart 46: Monthly values of transactions settled in InForex relative to the annual average 14,0 12,0 10,0 8,0 6,0 4,0 2,0 Inforex monthly turnover Inforex average yearly trunover

2004. január 2005. január 2006. január 2007. január 2008. január 2009. január 2008 szepte 2009 szepte It is distinctly visible in Chart 47 that examining the turnover of VIBER and InForex together, banks interbank transactions and the turnover of DVP items also declined, while the value of transactions concluded with the Central Bank increased sharply. The turnover of customer payments is even. Chart 47: Developments in individual payment order types of VIBER and InForex 90 000 80 000 70 000 60 000 50 000 40 000 30 000 20 000 other transactions customer payments DVP securities transactions transactions with the Central Bank banktobank items From the second half of 2008 the increase in the turnover of the MNB s twoweek bond introduced in 2007 had been restrained by the central bank deposit, then, following 2009 Q1, the bond came to the fore. The magnitude of endofday credit is negligible compared to the deposit/bond transactions. Chart 48: Transactions concluded with the Central Bank 20 000 18 000 16 000 14 000 12 000 8 000 6 000 4 000 2 000 Deposit with MNB MNB bonds O/N credit

Chart 49 shows the intraday distribution of the payment flows of VIBER and InForex together. The peak turnover following the opening is a result of the maturity of the central bank bond and the value date items. The surge after 9 o clock is caused by the settlement of the twoweek bonds. Chart 49: Intraday distribution of VIBER and InForex transactions in value by types including the MNB in December 2009 16 000 14 000 12 000 8 000 6 000 4 000 2 000 7 8 óra 8 9 óra 9 10 óra 10 11 óra 11 12 óra 12 13 óra 13 14 óra 14 15 óra 15 16 óra 16 17 óra before and after VIBER opening banktobank items customer payments hours DVPsecurities without MNB bonds MNB bonds transactions with the Central Bank ICS position matrix depozit with MNB "O"/N credit redemption The outstanding number of other items comprises the endofday credit/deposit transactions, interests and fees as well as the forint side of foreign exchange transactions. Chart 50: Intraday distribution of VIBER and InForex transactions in volume by types including the MNB in December 2009 peaces 30 000 25 000 20 000 15 000 5 000 7 8 óra 8 9 óra 9 10 óra 10 11 óra 11 12 óra 12 13 óra 13 14 óra 14 15 óra 15 16 óra 16 17 óra before and after VIBER banktobank items customer payments opening hours DVPsecurities without MNB bonds transactions with the Central Bank depozit with MNB MNB bonds ICS position matrix "O"/N credit redemption

1999. 2000. január 2001. január 2002. január 2003. január 2004. január 2005. január 2006. január 2007. január 2008. január 2009. január The intraday credit line requested by banks serves the purpose of supplementing the balance account. Its daily amount increased gradually (as a monthly average). The twoweek bond introduced in 2007 can also be placed as collateral, which resulted in a slight increase. A sharp increase was caused by the decline in the level of statutory reserves at end2008: it reached its peak (HUF 1 598 billion) in April 2009, followed by a sudden fall. 1 800,0 Chart 51: Monthly changes in the intraday credit line 1 600,0 1 400,0 1 200,0 1 000,0 800,0 600,0 400,0 200,0 The amount of the credit line can be modified in a discretionary manner during the day: it can be increased by providing additional collateral; the free stock can be reduced. The relevant intraday distribution in value for 2009 is shown in Chart 52. 2500 Chart 52: Intraday credit line amendment in value 2000 2 106,1 1 915,1 1500 1 205,2 1000 863,0 959,7 821,4 680,8 763,3 500 0 500 4,2 77,3 131,6 396,9 318,4 289,1 241,2 122,7 536,2 396,5 237,3 33,0 67,7 32,6 36,2 1000 78 89 910 1011 1112 1213 1314 1415 1516 1617 1718 1819 hours increasing decreasing

3. Payment transactions carried out by the Hungarian Post The growth rate of payments settled by the Post is determined by the steady increase in cash credit transfer, supplemented by the moderate increase in pensions by cash delivery order since 2001. Cash drawing by bank card represents a modest but increasing magnitude. Following the pattern of VIBER and the ICS a decline took place in the postal turnover as well by 2009. Chart 53: Turnover in value performed the Post by payment types 6 000 5 000 4 000 3 000 2 000 1 000 1998. 1999. 2000. 2001. 2002. 2003. 2004. 2005. 2006. 2007 2008 2009 Cash credit transfer Domestic postal order Cash payment with postal order Cash delivery order Pension postal order Banking card cash withdraw The number of transactions carried out by the Hungarian Post follows the trend shown in value. Chart 54: Turnover in volume performed the Post by payment types million peaces 350 300 250 200 150 100 50 1998. 1999. 2000. 2001. 2002. 2003. 2004. 2005. 2006 2007 2008 2009 Cash credit transfer Cash delivery order Domestic postal order Pension postal order Cash payment with postal order Banking card cash withdraw

Chart 55: Average amount per item in postal cash payments (1998 2009) HUF thousend/item 18,0 16,0 15,1 15,5 15,4 15,7 16,4 16,9 16,8 14,0 12,0 10,0 8,0 10,2 9,7 9,9 10,1 10,7 6,0 4,0 2,0 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 III. Comparison of the payment flows of individual systems The ratio of payment transactions settled by the GIRO and the MNB was almost exactly 5050 upon the launching of the interbank giro system (the turnover of the GIRO was 51.52% in value), then, following the launching of VIBER, from 2000 on most of the turnover was settled at the MNB; this ratio exceeded 94% in 2009. Chart 56: Distribution of the interbank payment turnover between the MNB s systems and the ICS 100% 51,52 49,25 46,11 47,49 43,78 33,43 23,19 14,74 10,70 9,96 8,55 8,63 6,70 6,29 5,90 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 1995. 1996. 1997. 1998. 1999. 2000. 2001. 2002. 2003. 2004. 2005. 2006. 2007. 2008. 2009. MNB (VIBER and InForex) ICS

2002. január 2003. január 2004. január 2005. január 2006. január 2007. január 2008. január 2009. január Following the launching of VIBER only approximately one third of the payment orders with an individual value exceeding HUF 10 million was settled in VIBER; this ratio has increased to above 50% since then. Charts 57 and 58 show the relevant annual and monthly developments. Chart 57: Annual distribution of customer payments exceeding HUF 10 million between the ICS and VIBER 100% 90% 80% 70% 30,3 30,3 31,9 41,3 43,3 46,5 52,7 50,7 60% 50% 40% 30% 20% 69,7 69,7 68,1 58,7 56,7 53,5 47,3 49,3 10% 0% 2002 2003 2004 2005 2006 2007 2008 2009 ICS customer payments exceeding HUF 10 million VIBER customer payments Chart 58: Monthly distribution of customer payments exceeding HUF 10 million between the ICS and VIBER 6 000 5 000 4 000 3 000 2 000 1 000 ICS customer payments exceeding HUF 10 million VIBER customer payments The turnovers in value of both VIBER and TARGET2 (which was launched in three stages between 19 November 2007 and 18 May 2008, and worked in parallel with the gradually terminated TARGET1 system) shows significant fluctuations; in several periods they move in the

Value settled per year (in billion ) same direction, and a declining trend is typical of 2009 (in spite of the increase in the number of TARGET2 member countries: Slovakia joined on 1 January 2009 /and Bulgaria as well in February 2010 as an out country/). Charts 59 and 60 depict the developments in the annual turnovers of TARGET and VIBER with the change compared to the turnover in the previous year. Chart 59: Annual payment flows in TARGET and changes compared to the previous year TARGET turnover turnover yearly increase 800,000 700,000 600,000 10.1% 9.1% 15.6% 10.7% 20% 15% 10% 500,000 400,000 300,000 5% 0% 5% 10% 200,000 100,000 19.3% 15% 20% 0 2005 2006 2007 2008 2009 25% (Source: TARGET ANNUAL REPORT 2009) Chart 60: Annual payment flows in VIBER and changes compared to the previous year HUF thousend bns 1 200,0 21,85 % 1 000,0 19,70 % % 25,00 20,00 800,0 15,00 10,25 % 12,65 % 600,0 10,00 400,0 5,00 200,0 2,44 % 2005. 2006. 2007. 2008 2009. VIBER turnover yearly increase 5,00

Daily turnover in billion (equivalent) Chart 61: Daily turnover of the largest highvalue payment systems by year 3,000 Major largevalue payment systems in the world CLS TARGET (EU) Fedwire (US) 2,500 2,000 1,500 1,000 500 0 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 (Source: TARGET ANNUAL REPORT 2009) Chart 62: Average daily VIBER turnover 4 500,0 4 000,0 3 500,0 3 000,0 2 500,0 2 000,0 1 500,0 1 000,0 500,0 2000. 2001. 2002. 2003. 2004. 2005. 2006. 2007. 2008 2009. VIBER daily turnover

2007. január 2008. január 2009. január 2007. január 2008. január 2009. január Chart 63: Comparison of the VIBER turnover with the TARGET2 turnover in value, by month VIBER 70 000,0 TARGET2 EUR billion 120 000 60 000,0 100 000 50 000,0 80 000 40 000,0 60 000 30 000,0 20 000,0 40 000,0 20 000 TARGET2 turnover value VIBER turnover value Comparing the two systems in terms of volume reveals a similar trend, but by end2009 some increase is typical of TARGET2. Chart 64: Comparison of the VIBER turnover with the TARGET2 turnover in terms of volume VIBER 70 000,0 TARGET2 EUR billion 120 000 60 000,0 100 000 50 000,0 80 000 40 000,0 60 000 30 000,0 20 000,0 40 000,0 20 000 TARGET2 turnover value VIBER turnover value

The following two charts show the intraday cumulated turnover distribution of TARGET and VIBER in terms of value and volume. Chart 65: Pattern of intraday flows 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Value 2008 Value 2009 Volume 2008 Volume 2009 Linear distribution 7 a.m. 8 a.m. 9 a.m. 10 a.m. 11 a.m. 12 a.m. 1 p.m. 2 p.m. 3 p.m. 4 p.m. 5 p.m. 6 p.m. (Source: TARGET ANNUAL REPORT 2009) Chart 66: Intraday distribution of VIBER turnover cumulated by value and volume in 2009 % 100,0 90,0 80,0 70,0 60,0 50,0 40,0 30,0 20,0 10,0 7 8 óra 8 9 óra 9 10 óra 10 11 óra 11 12 óra 12 13 óra 13 14 óra 14 15 óra 15 16 óra 16 17 óra VIBER value % VIBER volume %

The continuous increase that had lasted for years was replaced by a decline in the case of the Post in 2009, while the ICS was able to record some minimum increase. Chart 67: Annual turnover of the Post and the ICS in terms of volume million peaces 350 300 236,9 244,4 287,66 296,7 316,0 324,9 326,8 330,0 319,0 250 200 150 139,6 154,5 161,9 172,3 186,7 204,1 219,4 238,6 241,2 100 50 2001. 2002. 2003. 2004. 2005. 2006. 2007. 2008. 2009. Post ICS Developments in the turnover in terms of value are just the opposite: in 2009 the turnover of the ICS fell below the value of 2007, while the decline in the turnover of the Post was much smaller. Chart 68: Annual turnover of the Post and the ICS in terms of value 80 000 70 000 61 887 65 907 70 335 63 482 60 000 50 000 42 986 45 074 48 456 51 301 55 610 40 000 30 000 20 000 3 457 3 835 4 332 4 601 4 881 5 110 5 325 5 564 5 353 2001. 2002. 2003. 2004. 2005. 2006. 2007. 2008. 2009. Post ICS