Evolution AFORE

Counting

• The publicly mandated and privately administered pension system in Mexico was formally introduced in July 1997. The government did not want to decentralize the system of "head counting". It introduced a new method of counting people under the new regime mindful of the multiple account problems under the old SAR. This method plus mergers and acquisitions are described

BDNSAR

• A centralized national database system for pension information has been set up. This database is called the National Database for the Retirement System (Base de datos Nacional del SAR, BDNSAR). This has been set up with the specific purpose of having strict identification of persons matched with their money.

Problems with SAR92

• In the 1992 introduction of SAR accounts, the accounting process did not work at all. As a result, 10 million workers ended up with 65 million accounts

Banks as recipients

• IMSS entered an agreement with the thirteen largest banks in Mexico to act as "collecting entities". These collecting agencies use their bank branches to collect information and money from the employers. If the information received does not match some pre-established (and transparent) criteria, the payment is refused and the employer is notified. The quality of information therefore does not degrade over time.

CURP

• This method allows tracking of migration of persons across different funds as well as keeps track of movement across employers. It was recognized that the existing methods of unique identification through the numbers assigned by the IMSS or the Hacienda (called RFC) did not work. Thus, the government introduced a new (supposedly unique) identification number called CURP (Clave Única del Registro de Población).

PROCESAR

• To manage such a large database, a new database system was introduced. This entity called PROCESAR managed under BDNSAR. It is under strict government control but operated privately. It is supposed to keep track of all the individual accounts by establishing a one to one correspondence between the affiliates and the accounts.

PROCESAR

• For the integrity of the system, it is imperative that there is a regular flow of information and money in the system. We have described above the basics of checks and balances in the system. Here, we describe the timing of this process

Employer

Database (PROCESAR)

IMSS and INFONAVIT

Pension Fund

Flow of information

Bank (collecting agency)Day t

Day t+1

Day t+6

Day t+7Database (PROCESAR)

Fund ManagerCentral Bank

Day t+8

Day t+9

Figure 3.1

Employer

Central Bank

Fund Manager

Pension Fund

Flow of funds

Bank (collecting agency)

Day t

Day t+1

Day t+8

Day t+8

     

 

Flow of funds

• Suppose the employer successfully transfers the money to the collecting agency (bank) on day t. There are thirteen banks authorized to carry out these transactions. On day t+1, the banks transfer the money to Banco de Mexico (Central Bank). Exactly a week later (day t+8), the central bank transfers the funds to the fund managers. Fund managers credit the fund to the account holders of the funds on the same day (day t+8). This flow does not take into account the yield from the balance in the fund itself that has gathered over time. It only refers to the new flow of funds.

Flow of information

• The flow of information follows a similar path to the flow of funds. If the collecting agency (bank) gets the information on day t, it passes it on to PROCESAR on the following day (day t+1). PROCESAR gives the information to the fund managers on day t+6. Note that the flow of information takes place before the movement of money (which takes place on day t+8).

Flow of information

• Information is also sent to IMSS and INFONAVIT. Information is double-checked and the following day it is sent back to PROCESAR (day t+7). The information is sent to the Central Bank on day t+8. After the Central Bank verifies the information, it sends the money first to the pension funds on the same day (day t+8). Information is relayed to the pension funds on the next day (day t+9).

Duplication problem of AFORES

• Despite CURP, RFC, NSS there are problems of duplicate accounts

• In 2000, it was reportedly 500,000

• In 2005, it was over 1,000,000

• In 2009, it is over 1,500,000

• Industry: Agriculture, Construction

Number of funds

• In 1996, 44 funds registered, 2009 18 funds

Number of AFORES

17

14 13 1311 12 13

15

0

5

10

15

20

1998 1999 2000 2001 2002 2003 2004 2005

Affiliation

• The number of people in the new system rose from under a quarter of a million in February 1997 to over 15.5 million by the end of 1999. The growth has not been linear. In the first ten months, the number of affiliates grew at a tremendous pace until it hit about 10 million.

21,067,971

20,9

72,8

3520

,879

,507

20,6

24,8

0220

,759

,346

20,5

22,3

2220

,405

,759

20,3

24,7

5920

,197

,952

20,1

05,8

2920

,011

,144

19,9

20,7

38

19,7

21,4

5019

,827

,190

19,5

78,6

5919

,343

,498

19,2

71,6

39

18,4

49,3

77

17,6

25,2

97

17,1

96,9

1016

,999

,289

16,7

76,5

3216

,574

,425

16,3

80,1

5916

,198

,097

16,0

49,8

7515

,872

,985

15,7

11,7

9215

,594

,494

15,4

60,0

7915

,229

,608

14,9

88,7

9814

,874

,702

14,7

50,7

2114

,622

,217

14,4

63,1

1714

,332

,461

14,2

09,6

7114

,061

,706

13,9

39,2

7713

,827

,667

13,6

79,5

1413

,483

,212

13,3

05,5

9913

,073

,943

12,8

57,4

8412

,666

,341

12,4

51,3

6212

,112

,147

11,9

19,0

4411

,691

,780

11,4

42,4

5811

,188

,112

10,7

32,3

7410

,106

,926

9,33

5,41

47,

840,

630

6,55

7,30

74,

917,

569

232,

247

1,03

1,25

22,

103,

515

3,49

2,64

6

17,4

28,6

58

17,8

51,0

9018

,024

,876

18,2

27,7

36

18,6

62,2

3018

,865

,906

19,0

68,3

0419

,178

,232

2,29

4,52

0

2,65

6,66

7

3,02

2,08

3

3,26

1,03

2

3,71

7,66

8

3,98

6,29

3

4,22

6,61

5

4,49

1,74

4

4,69

1,39

3

4,89

3,11

2

5,11

1,29

4

5,38

3,78

2

5,85

9,83

2

6,07

2,86

1

5,89

2,85

8

6,21

7,43

5

6,44

4,22

3

6,57

2,59

6

6,72

0,62

16,

856,

058

7,03

1,34

57,

270,

743

7,39

8,42

27,

456,

020

7,52

9,86

9

9,385,926

8,9

80

,40

8

8,5

37

,80

6

8,2

47

,29

7

7,1

14

,46

1

7,0

51

,42

0

6,4

87

,36

0

0

5,000,000

10,000,000

15,000,000

20,000,000

25,000,000

Registrados Acum Asignados P. de S. Asignados a AFORE´s

1997 1999 2001

69,6

8795

,406

108,

590

40,4

62 89,0

07

121,

364

98,3

0914

2,33

379

,187

99,1

7396

,532

87,1

45

107,

065

84,9

12127,

700

124,

950

282,

260

109,

928

202,

398

214,

924

213,

292

222,

901

202,

931

173,

402

225,

793

231,

748

1,38

9,13

11,

072,

263

799,

005

232,

247

1,42

4,92

31,

639,

738

1,28

3,32

31,

494,

784

771,

512

625,

448

455,

738

254,

346

249,

322

227,

264

193,

103

339,

215

214,

979

191,

143

216,

459

231,

656

177,

613

196,

302

148,

153

111,

610

122,

429

147,

965

122,

790

130,

656

159,

100

128,

504

123,

981

114,

096 24

0,81

023

0,47

113

4,41

511

7,29

816

1,19

317

6,89

014

8,22

218

2,06

219

4,26

620

2,10

722

2,75

719

7,62

1

196,

639

0

200,000

400,000

600,000

800,000

1,000,000

1,200,000

1,400,000

1,600,000

1,800,000

Registrados Mes

The Good News

End Year Affiliates EAP Affiliates/

EAP

1999 15594503 39648333 39%

2000 17844956 40161543 44%

2001 26518534 40072856 66%

2002 29421202 41085736 72%

2003 31398282 41515672 76%

2004 33316492 43398755 77%

2005 35,276,277 41,880,800 84%

2006 37,408,828 42,846,100 87%

When did it happen?

Month Affiliates Growth

April 2001 18,657,474 1.16%

May 2001 18,865,906 1.12%

June 2001 25,555,664 35.46%

July 2001 25,665,592 0.43%

August 2001 26,297,659 2.46%

September 2001 26,353,396 0.21%

Big jump in numbers in June 2001: forced assignment of affiliates

Method of allocation of affiliates• CONSAR allocates affiliates by ranking all the

funds according to their charges equivalent over account balance for one year.

• It then takes the quartile of AFOREs with the lowest charges

• Thus, the fund with the lowest charge gets allocated 100 points. If the AFORE with the second lowest fees charge 80% of the lowest, then that AFORE gets 80 points and so on

• AFOREs are then allocated the new accounts based on their points

The Bad NewsContributors Affiliates Proportion

1997 7,769 11,188 69%

1998 879,979 13,827,674 64%

1999 948,855 15,594,503 61%

2000 10,379,823 17,844,956 58%

2001 11,864,672 26,518,534 45%

2002 12,292,152 29,421,202 42%

2003 12,577,265 31,398,282 40%

2004 12,751,029 33,316,492 38%

2005 13,557,086 35,276,277 38%

2006 13,919,377 37,408,828 37%

25%

30%

35%

40%

45%

50%

2,000,000 2,500,000 3,000,000 3,500,000 4,000,000 4,500,000 5,000,000 5,500,000 6,000,000

Proportion of affiliates contributing

Number of affiliates

Bigger funds have more affiliates contributing regularly

Appearance, Disappearance and Concentration

AFORE 2003 2004 2005

Allianz 3.92% * *

Azteca 1.42% 2.47% 2.84%

Banamex 17.87% 17.48% 16.93%

Bancomer 13.79% 12.98% 12.67%

Principal 10.22% 9.69% 9.63%

Profuturo 9.95% 10.09% 10.06%

Top 4 52.21% 50.63% 48.72%Share of top 4 has barely fallen

“Regular” contributors, 2004, 13mSalary No. of

contributorsPercent

0 to 1 667892 5.3%

1 to 2 4293589 33.8%

2 to 3 2572563 20.2%

3 to 4 1518439 11.9%

4 to 5 919184 7.2%

5 to 6 619877 4.9%

6 + 2118815 16.7%

60% of regular contributors earn < 3 minimum salary

Market share: Head count

• Limit set at 17% but on the number of affiliates

• Not on the investment

• What is the incentive for AFOREs?

Special Fund

• One fund was "special". The fund XXI (Siglo Veintiuno or Century Twenty One) was the privatized arm of the IMSS, the state-run pay as you go system. When this fund was being set up, many market analysts expected that it would be one of the largest funds. Mexican workers in the formal sector would instantly recognize IMSS brand name.

XXI

• The name IMSS was recognized by people, but it was associated with inefficiency. Therefore, it repelled more people than it managed to attract. It has less than 3.5% of the market share. This failure of the Mexican privatized arm of IMSS stands in sharp contrast with a similar effort in Uruguay.

República

• In Uruguay, the privatized state fund, República, managed to capture the largest market share there. By the end of 1999, República captured 37% of affiliates and 55% of total market funds. This fund was so successful for one simple reason. It had the backing of the government of Uruguay to insure the safety of the fund. In Mexico, the government did not issue any explicit guarantee for any fund.

How far?

• Suppose we assume that the growth rate of the AFOREs stay the same as the growth rate of the real GDP. If we assume that the real rate of return averages around 6%, then, in 2020, we could see the funds in AFOREs growing to around 40-50% of GDP. In Chile, the first twenty years of operation of the privatized mandated pension system generated funds worth 42% of GDP

Mexico vs. Chile

• In the first twenty years, the withdrawal from the system will be low. As the system matures, we will see a large outflow from the system as workers retire with substantial time in the privatized pension system.

• Mexico 6.5%, Chile 10% contribution

Investment regimes

• Right from the beginning, severe restrictions were imposed on the mandatory privatized pension funds in Mexico. The rationale was simple: The government did not want to take any risks that could jeopardize the faith in the system. Having credibility was important for the system. During the crisis of 1994-95, banks in Mexico were hit extremely hard.

Investment regimes

• The whole banking system had non-performing loans (basically it meant there was little hope of getting the principal back let alone interest due on these loans) to the order of 25-30% of total loan portfolios. The Federal Government in Mexico ended up assuming that loan. This meant that taxpayers in the end funded bad loans of the banks.

Where is the money?

• Largely in government bonds

• Smaller amount: private bond

• Problem of crowding out

• Notas estructuradas

• Foreign currency