HOME | ÍNDICE POR TÍTULO | NORMAS PUBLICACIÓN

Espacios. Vol. 37 (Nº 36) Año 2016. Pág. 32

Financial Performance Rankings as Trading Organizing Devices: The Case of Chilean Pension Funds

Rankings de desempeño financiero como dispositivos organizadores de comercio: El caso de los fondos de pensión chilenos

Juan Felipe ESPINOSA 1; José Ignacio ALARCÓN 2

Recibido: 10/11/16 • Aprobado: 15/11/2016

Contents

1. Introduction: Herding Behavior and Rankings at the Core

2. Literature Review

3. Methodology and methods

4. Analysis and Discussion

5. Conclusions

References

1. Introduction: Herding Behavior and Rankings at the Core

Economic literature is rich in evidence on herding and irrationality in the behavior of pension funds (see for example Radatz & Schmukler, 2011). In Chile, evidence of pension fund imitation and herd behavior is shown in several studies. For example, Castañeda and Rudolph (2010: 9) explain that pension systems regulators –like the Chilean ones- that demand minimum return guarantees -MRG- provisions to limit the risk of financial downturns motivate herding among Chilean pension portfolio managers. Furthermore, Muñoz (2015), following a principal-agent approach, explains that the relative financial performance ranking versus the performance ranking ‘per se’ leads to a superior moral hazard; that is, the acknowledgement that protection against risk could bring about new types of risks for the Chilean pension fund system AFP. All in all, Bravo and Vasquez (2005) show evidence of AFPs imitation strategy and they cannot deny the herd behavior indicator proposed by Lakonishok et al. (1992). Imitation and herding behavior is then at the core of the most important institutional investor of the Chilean capital market.

A fixed income investment manager that works for a Chilean pension claims that:

‘…Pension funds managers, traders, and analysts are always worried about the ranking. Everybody is looking at the rest of the industry…We are extremely careful when performing any operation, because we know that everybody is looking at us, and not only the rest of the industry but also banks, institutional investors and the rest’ (Pension Fund Fixed Income Investment Manager Interview)

Chilean investment professionals, and in particular traders –‘economists in the wild’, to use the famous expression by Callon (1998), put AFPs ranking at the center of their market analysis and their operations. Those experts-practitioners are talking about the ranking that shows the relative performance of the pension funds. These preliminary quotations show the importance of the relative performance of AFPs financial operations and their impact on the broad Chilean capital, and more in general, on the financial markets.
Because of its importance, the Chilean pension supervisor (www.safp.cl, SAFP) regulates the information that AFPs ranking offers about the performance of the pension funds. SAFP has a database of financial performance for each pension fund for the last month and the last twelve months plus the complete history since 2002 -where a major change affected the system. SAFP also offers financial performance since the date of the creation of the whole system, which was in 1981. The law ensures that pension funds companies do not use any other piece of data for advertising purposes or for users/clients information. Also, if the company informs the ranking position, they need to specify the period and the type of fund. Since 2002, there are five types of funds: A: The most risky; B: Risky; C: Intermediate; D: Conservative and E: More conservative, which are related to the informed ranking position (Berstein et al., 2011).

Regulation on AFPs rankings is relevant for the clients, funds and the society as a whole. As Muñoz (2015) has shown, AFP output ranking can lead to reduced pension funds’ performance just because of its moral hazard effects. Moreover, Bravo and Vasquez (2005) demonstrate that pension fund imitation strategy is linked to herd behavior. These economic accounts show how important economic reasons are in order to explain herd and imitation strategy in markets. However, what is hardly explained in these economic accounts is the mechanism and the complex chain of elements that allow the ranking to lead to financial markets herding phenomena. To fill the gap, this text offers a socio-material approach to herding mediated by market information in the form of prices and the pension funds ranking. Such socio-material approach explains the transmission mechanism between trading, pension funds ranking and Chilean financial markets.

2. Literature Review

2.1. Agential Aspects of Rankings

Chilean pension funds continually demands quantitative measures for its performance. Some relevant findings from sociology of measures and rankings show that such measures diffuse fast and have important arranging effects on human activities and institutions (Espeland & Stevens, 2008).  Espeland and Sauder (2007) claim that the influence of a ranking needs to be seen in terms of how is that this ‘device’ inspires ‘mechanisms of reactivity’ between actors in a particular market. In addition, rankings can be seen entwined with the linear simplification of complex relationships. In the case of this paper, rankings attribute value to the AFPs being observed (Espeland and Sauder, 2007; Espeland and Stevens, 2008). ‘Consequently, information appears more robust and definitive than it would if presented in more complicated forms’ (Espeland and Sauder, 2007, p. 17). All in all, rankings produce strong reactions of financial market actors.

The problem with evaluation is that it demands classification and then comparability. Additionally, Kjellberg et al. (2013, p. 13) comment that any valuation demands an ordering process. In Sorting Things Out (Bowker and Star, 1999) the authors show that in order to organize things, people need ‘a spatial, temporal, or spatio-temporal segmentation of the world’ (Bowker and Star, 1999, p.10 cited in Kjellberg et al. 2013, p. 13).

Following Foucault (1970), Bowker and Star claim that in modern times, and in particular within science, the central device that organizes activities is the ‘list’. It is the spatio-temporal form of the list the vehicle that scientists, researchers and academics use to produce world segments. Aligned with this necessary aspect of ordering for valuation, Beckert and Aspers (2011) claim that a ranking is an important judgment device. They offer the example of the wine market and universities to present a comprehensive understanding of how economic actors produce valuations of goods and services. Drawing on a varied source of authors, Beckert and Aspers (2011, p. 21) claim that a ‘ranking is a way of creating order out of what is offered in a market by reducing complexity to one single scale at the ordinal level’. What this paper is then trying to show is that the AFP ranking works as a ‘spatio-temporal segmentation’ that produces ordering within the clients of the funds and, as a consequence, within financial markets traders.

Following science and technology studies scholars Bowker and Stark (1999), it is possible to see that coordination comes from the classification and coding of the elements. Furthermore, Pollock and D’Adderio (2012) provide proof that markets are crucially affected by affordances (Gibson, 1977) and by any material and nonmaterial aspect of rankings. This last aspect of rankings moves the inquiry towards ‘further agential aspects that might extend’ the social analysis of rankings. In their paper, Pollock and D’Addeiro arrive to the conceptualization of ‘ranking device’. Pollock and D’Addeiro offer interesting remarks for the study of any type of ranking from a more agential–affordance and spatial-material- based approach.

2.2. A Performativity Approach to Herding

Stäheli (2006) shows a remarkable history of more than one century and a half of financial markets based on the idea of crowd and crowd thought. Rock (2006) explains that this phenomenon can be defined as economic actors following a crowd in a specified period of time. Advancing a more detailed distinction, Pitluck (2014, p. 6) defines crowding as ‘a trader behaving similarly to a previous trader, regardless of cause’.  Pitluck remarks that although econometrics –such as financial economic- has documented that crowding is a ‘pervasive’ phenomenon in financial markets, there are strong doubts about the relationship between crowding and herding. At the core of this author’s distinction, crowding herding is defined when the investor decision’s behavior is mediated by imitation. Non-herding crowding presents a similar behavior but without imitation.

Current behavioral approaches are returning to the concept of crowd psychology and crowd thought (see Borch 2007 and Hansen, 2015). Those approaches recuperate the figure of the crowd to economic discussions by the practitioner’s use of contrarian approaches based on the technical analysis of prices and traded amounts. Geographer Nigel Thrift (2006, p. 281) calls for ‘a semi-conscious process of imitation’ to understand present-day processes of capitalism. Borch (2007) explains that this affective dimension of imitation that explains crowd behavior is related to the idea of semiconscious suggestion that Gabriel Tarde posits in his texts about crowd thought.

Reviewing social studies of finance, Pitluck (2014) expresses that although studies like MacKenzie (2003, 2004 cited in Pitluck, 2014), Clark (2000, cited in Pitluck, 2014) and Beunza and Stark (2004, cited in Pitluck, 2014) have researched homogeneous behavior, these studies still leave open the question about imitation as the cause of crowding and crowding herding. Following Raafat et al. (2009), a convergent social behavior that could be seen in the market is still an open question for Social Studies of Finance and the performativity literature. Although this is an interesting question, the present text attempts to suggest that a much more important turn for this performativity literature is not only to study whether imitation is the mediation that explains herding based on crowding. Rather, a performativity economic sociology study of crowding and herding provides an insight into how crowd semantics might affect -and more interestingly, perform- financial market operations.

However, in order to study crowd and herding semantics from the point of view of the traders and operators of the market, it is important to follow the distinction that Vollmer (2016) draws about financial numbers –and by extension rankings too- as signs and signals. The author claims that numbers can perform the market. Nevertheless, such a solution to the problem of sign/signal delivers a new oppositional polarized solution where ‘numbers will be determined by certain aspects of reality or certain aspects of reality will be determined by numbers’ (Vollmer (2016, p. 32). That is why Vollmer calls for a distinction between signs and signals that offers a continuum instead of a harsh separation between them. That is why ‘a number that is a sign by virtue of referring to some slice of reality may also be a signal that makes people do something, and a number that is a signal will tend to rely on being understood as a sign to begin with’. This is a semantic approach to study this phenomenon that builds over Esposito (2013). Esposito develops a useful theoretical prop that offers an expansion of performative approaches towards financial markets. She calls for ‘a second-order observation of economics’. Esposito makes it possible to see the way this moral hazard phenomenon behaves; an issue observed by several economists in the Chilean pension funds industry can be explained by the recurrent observation that traders perform about others.

All in all, and using Pollock and D’Addeiro’s conceptualization, this text hypothesized that rankings are organizing devices. As a consequence, the central hypothesis is that pension funds ranking becomes an organizing device that could have the power to lead herding behavior within financial markets.

3. Methodology and methods

The objective of this paper is to illustrate the interconnection between the AFPs ranking and trading operations. In particular, the paper seeks to describe the relationships and connections between ranking, trading operations and herding behavior within the Chilean capital market, in other words, to consider the ranking as an ordering device by asking the question about how rankings shape the way that traders perform their operations. Within the research project, the specific subjects under study are those who make transactions and financial operations. These traders are known as ‘market makers’. Interviews led to produce conversations about the operations of this group of people allow the researcher to have a vantage point of the market, since this type of trader is located ‘at the center of the market who trades for their own account while providing continuous bids and offers to a diffuse population of investors and speculators located throughout the world' (Abolafia, 1996, p. 2-3).

Data production is based on interviews. In addition, interviews are informed by emerging inquiry from fieldwork observations. Specifically, sets of 20 in-depth interviews were produced for this study. Interviewees were mostly people who currently work, or have worked, in financial organizations in Santiago de Chile at the trading desks of three banks, two AFPs and one insurance company. Also, and complementing those interviews, the principal researcher of the project performed interviews to the Central Bank, the Ministry of Finance, Bank and Financial superintendent and to the Insurance and Exchange Superintendence officials. The interviews were carried out between September 2015 and January 2016 on a face-to-face basis. The information collected via interviews was analysed using an emergent and open codification (Benaquisto, 2008).

All in all, the methods used correspond to a qualitative content analysis approach (Cáceres, 2008).  The open and emergent codification process leads to the production of explorative aggregate codes based on mutual exclusion between them –exhaustive codification. The output of the process is thirty-five (35) different codes. In order to produce the codes and the analysis, the interviews were recorded and transcribed. The transcription process was conducted by social scientists who wrote ‘verbatim’ conversations between the interviewer and the interviewee. An important issue concerning the content analysis is the fact that interviews were performed in Spanish. The authors translated the quotations.

The analysis is primarily illustrated with samples of quotes from the interviews. Since this paper is still part of an ongoing research project, the results are not definitive and need to be read carefully. The imitations are mainly related to the access to AFP trading desks. Researchers had access to insurance companies and banks –including the Central Bank- trading desks but access to the AFP trading floors was denied.

4. Analysis and Discussion

4.1. From structural towards performativity issues

When Chilean pension funds buy or sell financial products, every actor of the market is watching their movements. As most of the traders and regulators claim, the information regarding their movements is followed by the rest of the actors with tremendous interest. From the point of view of economy, this is a structural issue within the Chilean pension funds system. The high level of operations plus the relatively minor capital market is a known issue that many economic studies offer as a structural explanation (see for example: Arrau & Valdez, 2002; Angostini & Budvenich, 2010; Larraín, 2003). Nevertheless, what could be much more interesting is ask how do the rest of market’s actors respond to such movements?

“The AFP is a long term investment and obviously makes lots of transactions because these are gigantic amounts of money but at the end other chaps are just intermediating… nothing else, they are receiving orders from thirds then they obviously need to be there plus these thirds have quite different objectives then shit you can find anything in there” (Trader 1, Investment Bank)

The trader is referring to pension fund administrators. He is aligned with previously cited economic studies, but at the same time, he adds an interesting point regarding trading and those who are following AFPs behavior. He refers to the others as intermediaries. Market makers, like the one speaking in the quotation, are intermediaries; such position allows them to see the market in a very different way than those who invest like an institutional investor. As Abolafia (1996) claims, they are traders who are situated at the center of the market and although they trade on their own (representing the bank), they provide bids and offers to investors and speculators.

This point is important because, although market makers still trade some proprietary funds of the banks, many of their operations have as an objective to ‘produce’ market, serving the bank client’s needs. Some of these clients are pension funds themselves. These funds are relevant bank clients and, at the same time, crucial information generators. Interestingly, market makers have this duality of objectives. On the one hand, they need to serve bank clients and on the other, they need to maximize their own book–portfolios. The information they gather from any pension fund ‘movement’ is at the core of their decision process:

“Yes they had been organizing a thing for two weeks and they arrived to a big position and then they give the price to the market… end of the situation… it is not like us because we can’t organize anything it is a day by day thinking crutches” (Market Maker Trader 2, Bank) Market maker discourse remarks that the price is signaling the market. It can be clearly seen in that discourse that the price – in Vollmer’s (2016) terms ‘a sign’- given by the AFP is relevant. From a more semantic point of view, the trader uses the expression ‘end of the situation’. This is interesting because it shows the power AFPs has over the market. Furthermore, the trader is making differences between the work at the bank and those who trade with institutional investors, such as pension fund administrators. Yet, what the traders present is aligned with the structural phenomenon already explained by economists.

However, these discourses show important semantic and communicative issues that build and rebuild market conditions recursively. In order to observe this, the discourse of an AFP’s actor is presented below:

“It’s made in different forms it goes managing but it is made gradually, any structural position it can’t be moved from one day to another you have to delay some days to move the structural position and obviously when you begin to move the more structural position you go out to sell a lot today and a lot tomorrow and the rest begins to understand that you are short and they try to arbitrate, then they are all pendant about what is that AFPs are doing because the flows are too high” (Fixed Income Trader, AFP)

At the core of this discourse, the AFP trader is recognizing the structural economic importance of their fund and of any movement they perform in the market, but he is also slipping in a point about the managing process of his operation. Price and quantity signs can be perfectly seen as signals, and such signals can be seen as performativity of the market (Callon 1998; MacKenzie and Millo 2003). However, that ‘structural performativity’ still explains crowding in economic terms. Furthermore, there is another set of economists claiming that there is a different explanation for herding based on crowding at the core of Chilean capital and financial markets, that is, the demand for minimum return guarantees -MRG- provisions that the law enforces to the funds (see Castillo, 1996; Pérez, 1988; Valdez, 1997, 2010; Zurita & Jara, 1997; Castañeda and Rudolph, 2010 and Bravo and Vasquez, 2005).

What we can also read in the discourse of the pension fund trader is his reference to self-reflexivity and, in particular, to second-order observation of his own effects on the market (in terms of Esposito, 2013). This is a movement that challenges purely structural economic explanations of herding based on AFPs market interaction. But who is observing AFPs and producing this circularity that Esposito claims is at the core of the economic and financial phenomena? Let us return to the bank trading desk in order to sort this question out. In the following quotation, a resident economist at the biggest bank’s trading desk in Chile explains his ideas about herd behavior, when being asked about it by the researcher:

“At the end, this regulatory frame I believe divides the field in a way that [they] behave this way I don’t know if this explains the totality of this behavior I believe that it is a significant variable that is still important… and the other is that at the end the AFPs move the market if one AFP knows I don’t know am that in front sidewalk they are doing this other thing of course I will consider it that is this is an important conditioner at this moment… I’m not talking about coordination but at the end of the time the information permeates” (Trading Desk Economist, Bank)

When the trading desk economist expresses that information permeates, he supports the idea of circular observation among market makers every time AFPs are in the market. In addition, he is using a metaphor that creates two poles –them and us- that almost all traders already enacted in their discourses. However, the economist shows certain ambiguity. On one hand, this trader intelligence expert is not completely accepting herding -based on the distinction made by Pitluck, 2014- because he claims that this could be a coordination problem among pension funds administrators, and at the same time, between them and us –the bank and other market operators. In particular, the economist aligns himself with the literature that explains the herding behavior within the regulatory frame (Law Decree 3500) from where MRG rules the actions of funds. If there is any performativity effect, it is the enactment of the law that divides the field and then allows institutional investors and banks to behave in a certain way –as the economist says. Furthermore, as Muñoz (2015) explains in detail, relative financial performance ranking versus the performance ranking ‘per se’ leads to a moral hazard within the entire Pension Fund Administration system. At this point, it is quite useful to follow Esposito (2014, P.6) who explains that the moral hazard concept was ‘originally developed in the field of insurances, [that] has now extended to every aspect of economic behavior, from the micro-economic level of contract theory and individual decisions to economic politics and finance’. A relative performance indicator can lead to poor results just because the rest of the pension funds administrators behave even more poorly. All in all, as Esposito (2014) claims, finance is always under the effects of moral hazard.

Using Esposito second-order observation approach, this study calls attention to a more careful consideration of the ‘others’ that the AFP trader presents in her discourse. A fundamental concept being used by the discourse of fund fixed income trader is that these others –like traders 1 and 2- are always trying to mediate among themselves. Following the performativity approach to arbitrage from Beunza and Stark (2004, p. 369), this operation demands a distribution ‘across persons and instruments as the trading room organizes interaction among diverse principles of valuation’. Following this distributed approach, a question emerges: What kind of mechanism allows the existence of herd behavior with sell-buy operations of financial products in these markets?

4.2. The Emergency of Herding Devices

Legal conditions that structurally explain herd behavior in Chile are well studied in the literature on pension funds public policy (see Hernández & Parro, 2005; Larraín, 2003; Valdez, 1997; Zurita & Gómez, 2003) and also in economic financial literature (see paper introduction). The above-mentioned authors tend to claim that the solution consists of freeing the prices and loosening the conditions of pension funds administrators, or to put it bluntly, demanding ‘laissez-faire’ approaches. Indeed, an important bureaucrat of the Ministry of Finance in charge of the general policies and regulations of the country capital market expresses this problem in the following quotation:

“What pension funds administrators are doing is to absorb because it is mandatory some of these resources and then investing maybe not in the best ehm investment alternatives that us why that is risky and ones need to be very careful on how one confronts this ehm and that’s why there is a segmentation because as they can invest in just one… just one part of the capital market some distortions are generated within this market“ (Ministry of Finance Capital Market Officer)

The officer is referring once more to the minimum return guarantees problem. As it is now clear from the previous section in this text, the law determined the selection of segments and products where AFPs are allowed to invest. The officer is establishing this mechanism as a segmentation device. For her, clients and funds understand the value of their operations based on this device. This is quite close to what Stark (2011, p. 324) has defined as ‘ordinal lists’. For Stark, an ordinal list is a spatio-temporal segmentation of the world. Stark explains that ‘we don't know what we are looking for but we can recognize it when we find it’ (Stark, 2011: 325). Considering the ranking as an ‘ordinal list’ allows its user to calculate the relative position of each type of fund and to organize the space of possibilities of the fund administration. From the point of view of the pension fund trader, the problem of the ranking is at the core of his work, as the quotation below shows:

“That is there were many times when we didn’t do good investments to cover the risk… many times it is frustrating to see that you can’t make an investment but the variable is risk and not just stability risk but other aspects like to be seen in the press that you made a bad investment and everybody is going to know about it imagine it my fund administrator company had a 30% of La Polar…” (Trader 2, AFP)

The trader’s discourse shows the difficulties of those who work investing in the trading desk of the pension fund. In addition, trader’s bonuses are being mixed up with the relative profitability of the fund but at the same time and as a consequence of the MRG problem, it is also entangled with the public image of the fund. It is exactly at this point where the sign becomes a signal (Vollmer, 2016) and everything the trader does is mediated not just by the regulation but also from those ‘others’ that look at their minute-by-minute operations. It is not just market makers looking professionally in a second-order ‘beauty contest’, to say quoting Esposito (2013), but also from a much more broader audience, one that is actually following the results of the pension fund management, the public that could or could not be a client of this particular organization.

5. Conclusions

AFPs ranking produces ‘mechanisms of reactivity’ that lead to expected and unexpected effects in traders operations and markets. As a consequence, AFPs ranking can be conceptualized as a mechanism that generates reactivity in those who use them. Such a mechanism enacts ‘intended and unintended’ consequences within the human activities of clients and, crucially for this paper, within many trading activities.

As economists demonstrate, pension funds results are constricted by law and more importantly, by financial products open to be purchased by the funds. Pension funds administrators sell the service of wealth management to those bound by law to save money for their retirement. The ranking exists to offer a solution for a central problem in the AFPs industry. Funds administration, as any service, needs to be evaluated. In order to evaluate AFPs funds, clients and traders need to classify the funds and then organize the results. In consequence, they need to classify and categorize. This categorization has effectuations on the entire financial market. At the center of this financial problem is Esposito’s (2014) idea that finance is always under the effects of moral hazard. The actor affects the world he is observing, but at the same time, he is also part of a system being under observation. As a consequence, there is a circularity that affects reality, producing unexpected issues that were previously non-existent.

Herding behavior is well supported by the interviewees of this study.  In fact, many of them build up instruments to produce sophisticated observations of the market. For example, some regulation officers commented that an important broker –who is related to one of the biggest banks in Chile- periodically produces reports about the flows of the pension funds. Other commentators expressed that some of them build up models that mimic the pension funds’ investments and simulate their strategies so they can see the effect they have in the market as a whole. However, as the social studies of finance have demonstrated, opportunism is still an important issue in trading. Arbitrage is an interesting strategy that market makers follow from their reading of pension funds going as a herd into the market.

Market makers at the bank trading desks enact strategies that reads from pension funds and their demands to the bank. Market makers come up with solutions for pension funds, but they also use any piece of information generated by themselves or their colleagues, like the inflation trader, who is located conveniently closely enough to the trading desk. The derivative trader spots opportunities and takes them. The market maker also uses information that he has got from the brokers’ proprietary ‘bid and ask’ over the counter operation platform electronic systems. All in all, market maker arbitrage strategy is distributed ‘across persons and instruments as the trading room organizes interaction among diverse principles of valuation’ (Beunza and Stark, 2004, p. 369).

Imitation among financial actors and, as the trader claims, to ‘go out together’ into the market, is at the base of this distributed arbitrage strategy. As a consequence, bank trading desks, insurance companies trading desks and others ‘go out together’ to mimic the strategy of AFPs. As Preda (2007) remarks, information is used and modified in trading activities to each trader’s advantage. These are ‘trading games’ that were described in detail by Abolafia (1996). Those games are part of ‘mechanisms of reactivity’. Such mechanisms are not just imitations but could be many different types of response trading strategies. Essentially, such games are framed by institutional aspects related to regulation –MRG- that is at the core of the coordination device being used by pension funds.

MRG and pension funds ‘ranking device’ (following Pollock and D’Adderio, 2012) become then a herding device for pension funds organizations and allow for the coordination of games/strategies for market makers. In any case, the ranking organizes the production of novelty in the market because, as Esposito (2013) claims, the effects of moral hazard are always present within financial markets, and more importantly, these second-order observations are productive by themselves. Crowds are now at the center of the financial markets again. However, this socialization still demands a heterogeneous network of devices like the one studied in this text. Focusing on the affordances and mediations that the ranking device offers gives the opportunity to understand performativity effects in a more nuanced way. Performativity is then a range, not a point. The ranking offers a number of possibilities. The ranking does not just perform a type of crowding and/or herding behavior but it is also an entrance door to the novelty in the Chilean capital market.

References

Abolafia, M. (1996). Making Markets: Opportunism and Restraint on Wall Street. Cambridge, MA: Harvard University Press.

Arrau, P. & Valdés, S. (2002). Para desconcentrar los fondos de pensiones y aumentar la competencia en su administración. Rev. Estudios Públicos Nº85 (invierno), pp. 77-99. Centro de Estudios Públicos.

Benaquisto, L. (2008). Codes and coding. The SAGE encyclopedia of qualitative research methods, Sage Publications, pp. 86-89.

Berstein, S., Fuentes, O., & Torrealba, N. (2011). Esquema de multifondos en chile (No. 43).

Beunza, D. and Stark, D. (2004). Tools of the Trade: The Socio-Technology of Arbitrage in a Wall Street Trading Room, Industrial and Corporate Change, 13, 369–400.

Bravo, D., & Vasquez, J. (2005). Minimum Return Bands and Herd Behavior in the Pension Fund Administrators. 1995 - 2003. Santiago.

Hansen, K. B. (2015). Contrarian investment philosophy in the American stock market: on investment advice and the crowd conundrum. Economy and Society, 44(4), 616-638.

Borch, C. (2007). Crowds and economic life: bringing an old figure back in. Economy and Society, 36(4), 549-573.

Cáceres, P. (2008). Análisis cualitativo de contenido: Una alternativa metodológica alcanzable. Psicoperspectivas. Individuo y Sociedad, 2(1), 53-82.

Callon, M. (1998a). An essay on framing and overflowing: economic externalities revisited by sociology. In M.

Callon, M. (1998b). The embeddedness of economic markets in economics. In M. Callon (Ed.), The laws of the markets (pp. 1–57). Blackwell Publishers, Oxford, UK.

Castañeda, P., & Rudolph, H. (2010). Portfolio choice, minimum return guarantees, and competition in dc pension systems. Documento de Trabajo N° 39.

Castillo, M. (1996). Modificación al sistema de pensiones: ¿fondos diferenciados? Rev. Estudios Públicos Nº63 (invierno), pp. 307-328. Centro de Estudios Públicos.

Clark, G. L. (2000) Pension Fund Capitalism, Oxford Geographical and Environmental Studies. Oxford and New York, Oxford University Press.

Espeland, W. & Sauder, M. (2007). Rankings and reactivity: How public measures recreate social worlds. American Journal of Sociology, 113(1), 1–40.

Espeland W. & Stevens M. (2008). A sociology of quantification. European Journal of Sociology. 49:401–36

Esposito, E. (2013). Economic Circularities and Second-Order Observation: The Reality of Ratings. Sociologica, 7(2), 1–10.

Esposito, E. (2014). Circularidades económicas y observación de segundo orden: La realidad de las calificaciones crediticias. Revista Mad, (30), 1-24.

Foucault, Michel. 1970. The Order of Things: An Archaeology of the Human Sciences. London: Tavistock Publications.

Gibson, J. (1977). The concept of affordances. In R. Shaw & J. Bransford (Eds.), Perceiving, Acting, and Knowing: Toward an Ecological Psychology (pp. 67–82).

Hernández, L. & Parro, F. (2005). Sistema financiero y crecimiento económico en Chile. Rev. Estudios Públicos Nº99 (invierno), pp. 97-138. Centro de Estudios Públicos.         

Kjellberg, H.; Mallard, A.; Arjali, D-L.; Aspers, P.; B., S.; Bidet, A.; Corsin, A.; Didier, E.; Fourcade, M.; Geiger, S. and others (2013). Valuation studies? Our collective two cents. Valuation Studies, vol.1, issue 1, p.11-30.

Lakonishok, A., Shleifer, A., & Vishny, R. W. (1992). The impact of institutional trading on stock prices. Journal of Financial Economics, 32(1), 23. http://doi.org/10.1016/0304-405X(92)90023-Q.

Larraín, C. (2003). Supervisión financiera: lecciones del caso Inverlink. Rev. Estudios Públicos Nº92 (invierno), pp. 111-142. Centro de Estudios Públicos.

MacKenzie, D. (2003). Long-Term Capital Management and the Sociology of Arbitrage, Economy and Society, 32, 349–380.

MacKenzie, D. (2004). Social Connectivities in Global Financial Markets’, Environment and Planning D: Society and Space, 22, 83–101.

MacKenzie, D. & Millo, Y. (2003) Constructing a Market, Performing Theory: The Historical Sociology of a Financial Derivatives Exchange. American Journal of Sociology 109: 107-145

Muñoz, G. (2015). Evaluacion de Desempeño de AFP’s: Una aproximación para evitar costos asociados al Riesgo Moral. Pontificia Universidad Católica de Chile.

Pérez, F. (1988). Necesidades de inversión para los Fondos de Pensiones. Rev. Estudios Públicos Nº29 (invierno), pp. 171-303. Centro de Estudios Públicos.

Pitluck, A. Z. (2014). Watching foreigners: how counterparties enable herds, crowds, and generate liquidity in financial markets. Socio-Economic Review, 12(1), 5-31.

Pollock, N., & D’Adderio, L. (2012). Give me a two-by-two matrix and I will create the market: Rankings, graphic visualisations and sociomateriality. Accounting, Organizations and Society, 37(8), 565–586.

Raafat, R. M., Chater, N., & Frith, C. (2009). Herding in humans. Trends in Cognitive Sciences, 13(10), 420–428.

Radatz, C., & Schmukler, S. (2011). Deconstructing Herding: Evidence from Pension Fund Investment Behavior (No. 5700).

Rock, L. (2006). An economic psychological approach to herd behavior. Journal of Economic Issues, 40(1), 75-95.

Stäheli, U. (2006). Market Crowds. In Schnapp, J. T. and Tiews, M. (eds) Crowds, Stanford, CA, Stanford University Press, pp. 271 –287.

Stark, D. (2011). What is Valuable? In Patrik Aspers and Jens Beckert, Eds. The Worth of Goods: Valuation and Pricing in the Economy. Oxford: Oxford University Press.

Valdés, S. (1997). Libertad de precios para las AFP: aún insuficiente. Rev. Estudios Públicos Nº68 (primavera), pp. 126-147. Centro de Estudios Públicos.

Vollmer, H. V. (2016). Financial numbers as signs and signals: looking back and moving forward. Economic Sociology the European Electronic Newsletter, 17 (2). 32-38

Zurita, S. & Jara, C. (1999). Desempeño financiero de fondos de pensiones. Rev. Estudios Públicos Nº74 (primavera), pp. 227-254. Centro de Estudios Públicos.

Zurita, S. Gómez, L. (2003). Normativa de los mercados derivados en Chile. Rev. Estudios Públicos, Nº89 (verano), pp. 63-90. Centro de Estudios Públicos.

1. Facultad de Economía y Negocios de la UNAB. Email: juan.felipe.espinosa@gmail.com

2. Facultad de Ciencias Sociales, Universidad de Chile

3. Proyecto Fondecyt N° 11140432: The Relevance of Crowd Though in the Chilean Capital Market