Invested interests? Reflexivity,representation and reporting in financial markets

Introduction

Money may make the world go round but communication systems are needed to make real-time flows of capital go round the world. They also provide the interconnecting infrastructures and real-time information systems that link exchanges, trading rooms, and investment portfolios. Insofar as these systems increase the informational efficiency of financial trading, translating new information into prices in real time, financial information networks also provide the interconnections through which bubbles and panics can proliferate. Although financial media are not themselves responsible for crises, the need to understand the role they play in finance has never been greater.

The scale of the global ‘credit crunch’ crisis in 2007–8 resulted in both an unprecedented implosion of financial asset values and an equally unprecedented level of government intervention to stabilise the banking sectors in the USA and EU. Continuing financial uncertainties in the EU are indicative that the ramifications of the 2008 crisis are still unfolding. The potential for further political and economic turmoil places the financial news media in a position of significant responsibility. There is a pressing need for policy-makers or regulators, market actors and the general public to understand the causes of, and potential solutions to, the current fragility of the financial system, and to appreciate the implications of different policy responses. However, financial markets are complex phenomena which resist simple definition and explanation. The expansion of financial market activities since the 1970s must be understood partly in terms of the macroeconomic shift from the Bretton Woods arrangement to monetarist models emphasising free capital flows and tight domestic fiscal policy. However, equally significant was the emergence of new communication/information technologies (including satellite links, computing, and real-time data networks) which facilitated the development of complex new financial instruments and computational models for investment strategies (Thompson, 2003).

As Wayne Parsons’ historical analysis of the financial media in the UK (1989) demonstrates, the evolution of financial markets is inextricable from the evolution of the news and information systems that support their functioning and also shape policy discourses and the attitude of governments and the public toward finance (see also Cheney, 1998; Greenfield and Williams, 2007; Thrift, 2001). The expansion of financial market activity has gone hand-in-hand with the development of increasingly sophisticated real-time financial information and news services. Central to these are the major financial wire services, notably Thomson-Reuters and Bloomberg (Bartram, 2003; Craig, 2001; Palmer et al., 1998). These provide both real-time electronic news feeds and the trading platforms which link trading rooms and exchanges around the world. Financial transactions worth trillions of dollars flow on a daily basis through these communication networks (Bank of International Settlements, 2011). The growth of financial markets and increased opportunities for the general public to actively manage their own investment activity have been reflected (and in part, driven) by the concomitant attention paid to the financial markets in general news media, online financial chat-rooms, and specialist channels such as CNBC (see Greenfield and Williams, 2007; Thompson, 2010c).

As institutions, the financial media and the news services they provide cannot be regarded simply as external and independent observers of market events. These media are an intrinsic component of the financial system, but this means that the relation between financial reporting and financial trading activity is far from straightforward. Several studies have suggested that the financial news can exert an influence on trading and market prices, at least in the short term (e.g. Busse and Green, 2002; Sant and Zaman, 1996; Vickers and Weiss, 2000). As Aeron Davis (2005, 2006, 2011) suggests, any notion of media ‘effects’ on the market must take account of financial traders, analysts, and fund managers who engage with the media as an active audience with expert knowledge. Nevertheless, there is evidence that financial reporters are often dependent upon elite sources, especially investment bank analysts who often act as the primary definers of financial events (Golding, 2003; Hong and Ki, 2007; Kurtz, 2000; Thompson, 2009).

Drawing on the author’s doctoral research into media usage by financial traders and analysts based in New Zealand, the article explores the relation between financial journalism and financial market activity and problematises the reflexive, constitutive nature of information in markets. The first part problematises neoclassical notions of representational market efficiency and outlines alternative conceptions of informational reflexivity. The next section considers the implications of this in respect to the relationship between financial media reporters and their market sources and the potential for this relation to manifest in reports that reflect and reinforce market consensus. This is then followed by a discussion of findings from interviews with finance professionals about their sourcing of information and relations with financial journalists.

Information and market reality

Timely and accurate information is essential to the operation of financial markets. Trading rooms across the globe are linked into real-time financial news services which also provide the network infrastructure for electronic brokerage systems through which the majority of trading activity takes place. The neoclassical conception of efficient markets supposes that information is automatically and instantaneously incorporated into prices (Fama, 1970). Concomitantly, the more information that is made available, the more prices will be a valid indication of fundamentals and any discrepancies will immediately be resolved by traders seeking arbitrage opportunities. On a basic level, the instantaneity and symmetry of real-time financial information systems confirms such principles, and trading responses to market news rapidly incorporate new data into prices. However, the conception of markets assumed here is problematic in its supposition that information provides a more or less accurate representation of an objective, external market reality.

Crucially, this positivistic, representational conception of market information does not take account of the symbolic ontology of finance capital and the dependence of market values on investment models and the frameworks of interpretation currently deployed by investors. The neoclassical perspective tends to attribute financial risk and instability to extraneous non-market inefficiencies of insufficient information, hence the emphasis on transparency and disclosure in many financial regulatory regimes. This misconstrues the extent to which financial risks are endogenously generated by the collective expectations of investors (see Best, 2005; Bryan and Rafferty, 2006; also Minsky, 1977). As George Soros (1994, 2008) has suggested, the relation between fundamentals and prices is not simply representative, but reflexive. As prices respond to new information in real time, the price changes register on trading screens, feeding back into investor perceptions. Insofar as market prices reflect the aggregate perceptions of the investment community, the truth value of financial information is not necessarily independent of the extent to which it is collectively believed and traded on. This poses an interesting epistemological challenge for financial journalists, because the nature of financial markets precludes their direct observation as a publicly accessible event (as opposed to, say, a traffic accident or a protest). Verification of financial events therefore requires some form of mediation, either through the presentation of market data on screens or through articulation by expert primary definers within the markets (Thompson, 2009). In turn, financial news feeds back into the trading rooms and helps inform trading decisions.

This suggests that financial information systems are not representative of independent market conditions but immanently constitutive of them. A range of studies on the sociology and anthropology of finance supports such a contention. For example, Beunza and Stark (2004: 370) contend that ‘electronic markets constitute an on-screen reality that lacks an off-screen counterpart’. Knorr-Cetina and Bruegger (2002a, 2002b) and Knorr-Cetina and Preda (2007) have similarly suggested that trading screen displays should be conceptualised through the phenomenological notion of ‘appresentation’, in the sense that the data directly constitute the financial environment. As Knorr-Cetina and Bruegger argue, ‘Market reality […] has no existence independent from the informational presentation of the market on screen that is provided by the news agencies, analysts, and traders themselves’ (2002b: 915). The appresentation of this constructed reality is therefore constantly in the process of being reproduced, and is therefore ontologically distinct from material/spatial market reality.

Thompson (2003, 2010b), meanwhile, identifies three basic forms of communicative reflexivity in financial markets:

In some circumstances, game reflexivity may also extend to ‘herding’ behaviour and self-fulfilling prophecies whereby large numbers of investors become sensitive to shifts in market sentiment, particularly during periods of volatility or transition points in a cycle where an upturn/downturn is expected. However, not all correlated trading behaviour reflects herding, and in most circumstances this would not be an intentional trading strategy among most major investment institutions (Bikhchandani and Sharma, 2001). As Hyman Minsky’s Financial Instability Hypothesis (1977; see also Kindleberger, 1996) recognised, periodic cycles of market bubbles and panics reflect the evolution of market expectations so as to engender collective overconfidence and the discounting of risk during periods of growth, followed by an over-reaction and ‘rush to the exit’ when a shock (such as a major default) triggers a revision of market valuations. Importantly, the expansion of market values encourages expectations of further growth. As Minsky (1977) points out, this often encourages cheaper credit and borrowing to leverage trading positions, thereby creating the conditions for a subsequent collapse when the expansion of values cannot be sustained. Thus the expansion of fictitious values constitutes an endogenous signal to the market, ostensibly confirming that the prevailing market consensus underpinned by current valuation models or trading frames is valid. However, once the consensus is shattered, the intersubjective coherence of the valuation models breaks down and fictitious values implode as everyone tries to offload undesirable assets into an illiquid market (especially when positions are highly leveraged). This is what occurred in the credit crunch, particularly after the collapse of Bear Stearns and Lehman Brothers and the ratings agencies’ downgrading of an entire class of mortgage securities to junk status (Thompson, 2010a, 2010b; Winseck, 2010).

It is therefore important to recognise the reflexive communicative processes which underpin market bubbles and crises. The intersubjective codes that define what counts as a ‘fundamental’ in trading frames and permit asset values to be calculated coherently can periodically be subject to performative redefinition. Applied to Minsky’s model of instability, the reflexivity processes outlined above help explain how investors become sensitised to the signals their own collective behaviour generates, and hence why the consensus view during the former phase frames the expansion as a sign of robustness, with the latent fragility being recognised only after the onset of crisis (Thompson, 2010b).

Several researchers have suggested links between the financial media and self-referential feedback loops in markets (Bryan, 2001; Davis, 2011; Graham, 2006; Kunczik, 2002; Rothkopf, 1999; Thompson, 2010a). Nevertheless, the arguments concerning appresentation and reflexivity should not be taken either as a generalised assertion of radical autopoiesis whereby contemporary finance has become completely disconnected from the ‘real’ economy, or as a claim that irrational herd-behaviour in pursuit of self-fulfilling prophecies is characteristic of financial investors. Over the long term, core market fundamentals (such as supply, demand, price data, ask-bid spreads, company earnings reports, and ratings agency grades) remain the enduring variables around which asset prices generally fluctuate. ³ However, the ‘irrational exuberance’ (Greenspan, 1996) evident in the disproportionate expansion of financial asset values relative to global GDP (financial depth) ⁴ since the 1980s is indicative that the shared meanings and valuations ascribed to these factors have evolved.

Neoclassical conceptions of market efficiency and rationality must therefore be regarded as suspect, but important questions about the precise role the financial media play in these processes remain. During periods of market uncertainty, mutual monitoring among market actors is likely to intensify as traders seek to detect signs of shifts in market consensus and anticipate how investors will collectively respond to forthcoming market news announcements. However, the notion of ‘consensus’ here does not imply absolute agreement across the market. Indeed, if all investors had exactly the same valuations and expectations about assets and price movements, trading and values would be disrupted (which is precisely what occurs in crises where investors rush to offload undesirable assets into an illiquid market, precipitating the anticipated collapse in prices). Symmetrical distribution of information therefore does not imply concomitant symmetry in meaning or trading response (Thompson, 2003). Market consensus can be better understood as the prevailing set of intersubjective codes which demarcate the broad parameters within which investor expectations will normally vary over a market cycle. So long as price movements remain within that range the validity of the prevailing investment models, frames, or schemata are confirmed, but when price movements violate the consensus, then the underlying assumptions require renegotiation. The complex relation between journalists and the institutional sources who define market consensus and the role of the financial media in communicating shifts therein is addressed in the following section.

Financial media and source–reporter relations

Financial media play a role in providing a constant indication of market consensus among analysts and investors (Davis, 2005). The analysts of investment banks and other major financial institutions are typically the primary sources whom financial reporters routinely consult for definitions of market events. However, the recognition of reflexive information processes poses both a theoretical and practical problem for financial reporting. Markets can never simply be accessed and represented as a set of objective, verifiable facts. Theories of media representation have long recognised that news is a social construct, not a simple mirror of reality (e.g. Manning, 2001; Tuchman, 1978). Even ostensibly concrete market facts such as price changes on exchanges, company earnings, or ratings agency upgrades/downgrades require interpretation and contextualisation. Thus their truth value is not independent of the ongoing performance of the theories, models, and practices which define them as real. In a market environment characterised by real-time information flows, many investment strategies demand very rapid decisions in response to unfolding trends. This increases the pressure on information providers, including news media, for continuous and rapid updates. During periods of uncertainty or crisis, the pressures for immediacy will often be accompanied by a break-down in the intersubjective recognition of models and valuation criteria that significantly complicate efforts to define unfolding events.

It is likewise difficult for reporters to make sense of financial market activities outside of the intersubjective codes and epistemologies provided by financial actors. Some experienced reporters and commentators in the more specialised financial media do have expert knowledge (and sometimes professional investment experience). However, the complexity and specialisation of contemporary financial theory, investment instruments, and trading strategies usually means the reporter is often not in a position to contradict the source except by reference to other authorities (Lenzner, 1997; Matolcsy and Schulz, 1994). Consequently, journalists rely on analysts who have the epistemological authority to define market reality. These definitions are contestable within the prevailing paradigm of professional finance, but routine financial news production generally has to accept the premises of the sources in order to make sense as financial news. As Starkman observes:

The business press exists within the Wall Street and corporate subculture and understandably must adopt its idioms and customs, the better to translate them for the rest of us. […] It is far easier for new bureaucracies to accept ever-narrowing frames of discourse, frames forcefully pushed by industry, even if those frames marginalize and eventually exclude the business press’s own great investigative traditions. (2009)

Moreover, compressed news-cycles and the valorisation of real-time reporting means news workers have less time for analysis and corroborating sources, increasing the risk of ‘capture’ by experts who can provide information subsidies in the time-frames newsrooms require (Aronson and Sylvie,1997). Although the 2008 crisis opened up space for questioning the banking system and the ostensible hazards of exotic mortgage securities in regular news media coverage and some editorials (Thompson, 2009), the financial news continued to privilege the voices of financial experts. Notable exceptions include Michael Hudson (2007) and Gillian Tett (2007), who critically reported the problems in the mortgage securities sector before the crisis, and high profile finance figures such as Warren Buffett (2003), George Soros (2008), and George Stiglitz (2010), who have called for reforms. Nevertheless, calls for re-regulation have tended to limit the scope of legitimate criticism by diagnosing symptomatic elements of financial market dysfunctions rather than their systemic origins (Thompson, 2009). As Starkman comments in his critique of the financial media’s coverage of the sub-prime crisis, ‘There does seem to be a tendency in big financial newsrooms to zoom in on esoteric stories on the margins […] and ignore the big, dumb, honking ones at the heart of the financial system’ (2008: 50).

The relationship between financial experts and journalists has significant implications for policy-makers as well as investors. Government responses to the unfolding crisis were compressed into time-frames that made definitive assessment of the risk and cost associated with policy responses impossible. Whether or not government bail-outs of the banking sector restabilised the credit system was partly contingent on the extent of the investment community’s collective confidence in their efficacy. Although technical charts can map patterns of risk historically, the ontology of financial risk in the unfolding present is inseparable from the models used to performatively calculate it (Best, 2005; Bryan and Rafferty, 2006). Indeed, as Hope (2011) points out, the growing complexity and scale of finance since the 1990s has seen financial policy-makers becoming increasingly reliant on specialists and advisors from the finance sector. The implication here is not corruption, but an institutional transcoding of the epistemological frameworks, discourses, and norms underpinning financial investment practices into the policy sector. Sebastian Schich (2010), principal administrator of the OECD Financial Affairs Division, acknowledged that the policy frameworks needed to guide intervention as the crisis evolved were themselves unclear. This is doubly problematic in a crisis when the investment community is likely to be highly sensitive to the responses of government. In such scenarios, the default frameworks for policy-makers are liable to be inflected by the epistemologies of the institutions they are regulating, even when they themselves are approached by the media for authoritative definitions of financial events.

Crucially, the expert sources the financial media routinely consult often have (in)vested interests. Analysts are often institutionally biased in their public media statements and investment recommendations (Beunza and Garud, 2005; Hong and Ki, 2007), and those who work on the ‘sell-side’ for investment bank clients are often complicit with the priorities of investor relations (Golding, 2003l; Kurtz, 2000). Indeed, the institutional pressures on analysts are in many respects comparable to the ‘propaganda model’ filters of news production (Thompson, 2009). Collectively, analyst comments in news media provide an index of current market consensus on market expectations on a range of markets and individual securities (Davis, 2005), although there is often a difference between the publicly reported consensus and the private ‘whisper numbers’ that are communicated through institutional networks (Kurtz, 2000; Thompson, 2010c). The price sensitivity of a company’s earnings announcement, for example, typically depends on the consensus expectations among market analysts, and a negative surprise can drive share prices down sharply. Investor relations efforts target consensus expectations, and pre-announcement news releases might hint at weak earnings performance so even a poor result is framed in relatively positive terms (Golding, 2003; also Davis, 2007; Graham, 2006).

Although such tendencies are well recognised by professional traders, the analysts who push these frames on behalf of clients are nevertheless an important source of information in financial markets. On trading floors, analysts help provide frames of reference that allow traders to more rapidly ascribe significance to new information in relation to investment schemata. Rothkopf (1999) and Kunczik (2002) have both identified an institutional predilection for self-referentiality among financial analysts and reporters. The routinisation of financial news production, investor relations efforts and public analyst commentaries can promote a symbiotic relationship among these actors. As Rothkopf comments:

Who does the media turn to for analysis? Experts. And who are the experts? They are a very limited group of academics, Wall Street executives, former government officials and media commentators. When one reads an article or watches a television interview programme, one hears primarily from a small coterie of familiar faces – and hears from them over and over again. […] Thus, while the media seems increasingly diverse, its sources are remarkably homogenous and close-knit. (1999: 89–90)

Investor–reporter relations – empirical evidence from New Zealand

The following section provides some further insights into the relationship between journalists and analysts, drawing on the findings from the author’s (pre-crisis) research into media and information usage by financial professionals. ⁵ This included 39 semi- structured interviews with New Zealand-based traders and analysts, and several periods of non-participant observation in the trading rooms at Deutsche Bank, ANZ, and the Reserve Bank of New Zealand. Two local financial journalists with prior experience at Reuters and the Financial Times were also interviewed.

The trading floors within the investment institutions studied were arranged into various desks dealing in different assets. Interviews suggested traders were relatively autonomous in routine trading decisions, although their institutional parameters for overall risk exposure and losses were managed centrally. The resident analysts provided daily briefings on particular market trends and were particularly active in analysing expectations in the run-up to scheduled announcements such as Reserve Bank interest rate (official cash rate) announcements. The analysts reported being routinely consulted to provide contextual interpretations of emergent market trends and help traders assess whether price movements were short-term ‘noise’ or indicative of changes in market conditions that required portfolio adjustment. Public media engagement was generally regarded as professionally and institutionally beneficial, partly because cultivating a media profile afforded them symbolic authority, but also because being a recognised primary definer allowed the analyst the opportunity to promote a preferred frame of explanation in times of market uncertainty.

A key component of this process mentioned repeatedly by respondents was the importance of discerning shifts in the current market drivers and the development of stories or frames that made sense of the incoming data in ways that made trading decisions meaningful. As one senior bank trader observed:

The art of trading is deciding what’s the flavour at the time, you know. What’s moving the market at the time? There’s certain data that moves the market, and certain data that doesn’t. It all depends what’s the flavour at the time. And that’s what happens, you know – Is it whether interest rates are moving currencies at the moment, or whether current account deficits are moving currencies? – It all depends. (Trader A)

Influencing this kind of meta-information can be advantageous for traders and analysts because if their preferred stories or frames are picked up and circulated through institutional networks and the financial media, the rest of the market might begin to align itself with their expectations. As one senior trader with a sell-side role confirmed, sometimes the stories/themes being promoted circulate to the point of becoming self-referential, but there are often competing counter-themes being pushed by rivals:

Part of my role is a sales role, so I can talk to all the major accounts in New Zealand. But I might wake up with a particular theme in my mind, and of course I’ll ring everyone and push that. […] If they like that idea, they’ll say ‘that’s a great idea’ – So a rival bank will ring up and I’ll say, ‘Yeah, but what about this?’ – as if it’s their idea! Then I’ll hear it coming back to me, and suddenly that might be the theme in the market. And it’s something I sort of woke up with in my head. (Trader B)

Another senior bank trader commented that there was a ‘massive competitive advantage’ in having one’s institutional analysts ranked highly for their global financial research. Quite apart from informing trading, there were other advantages in cultivating a media profile to ensure news releases would be picked up across a range of news media:

If you’ve got good relationships with Dominion, Herald, Dow Jones, Reuters, you can send them your new piece and they’ll give it some coverage […] Because you want the credibility, you want the profile, and you want the influence on the market, [and] the RBNZ. (Trader C)

This was confirmed by a senior analyst who pointed out that a media profile helped improve institutional credibility both with other market actors and potential corporate clients: ‘You want top business people to think we’re a player in the markets. We want our view to be known to the people who are making decisions – like the Reserve Bank, for instance’ (Analyst A).

Several interviewees pointed out that sell-side analysts and brokers would always look at incoming news to try and spot new angles and ‘create stories’ to promote trading. In other words, the aim was to identify a trading frame or narrative which would highlight certain events or factors ostensibly moving the market and provide a rationale for taking a particular position. As another senior bank analyst observed, there is a persuasive aspect to this:

Your views are a portfolio – you might have a dozen views. Two will be outstanding, two will be dogs. And hopefully the other half dozen will carry the day. […] there are myriad styles for investors […] So what you’ve got to try and do is, from all the information that’s available, try and [identify] a story – and then use that story to generate business. (Analyst B)

With the exception of in-house analysts advising their own colleagues, it would be unusual for professional traders to respond uncritically to analyst commentaries either in the news media or through networks of professional contacts. The perceived utility of such comments in the news media largely depends on whether their stories, frames, or schemata aligned with those currently assumed by traders:

If you rolled in the analysts, they’ve got no perspective […] they’re no bloody good […] If you took any analyst, you can ask [a trader] – all that economic information and the e-mails you’re given – Is it valuable? And he’ll say ‘Shit, no – I get bloody e-mails from 40 analysts. I only read 2 or 3’ – Why do you read two or three? – ‘Oh, I find those really good because I’m on the same wavelength as those guys and they help me make my trades’ – So why don’t we fire the other 37? Well of course, the other 37 are hitting someone else. (Trader D)

Other analysts doubted there was any individual commentator in New Zealand with sufficient influence to move the market (although it was noted that this is more likely in larger financial centres like Wall Street). However, several respondents did cite a case where a high-profile analyst triggered a flurry of trading activity by reversing their previous (consensus) view on interest rate policy, creating the perception that they may have acquired privileged information ahead of other market actors. As one senior bank analyst observed, if a high-profile opinion-leader went public with a statement challenging consensus opinion, other market actors might start to second-guess whether they were privy to new price-sensitive information:

I can remember times when the markets have moved a couple of points – just a small movement – when someone’s had a view which seems a bit strange. […] So if someone had recently seen the governor of the Reserve Bank and then suddenly has a radical change of view, you think, ‘hang on – what’s going on here?’ […] There might be a market reaction because people in the market might conclude that there’s a risk that there was […] some information in what that analyst was saying. (Analyst C)

Interviewees suggested that it was unlikely that financial journalists in New Zealand had the clout to move the market (although again, this was regarded as possible in other centres). Several respondents remarked on cases in Australia where stories by journalists perceived to have close connections to key institutions had triggered market reactions.

You can never categorically say that it’s happened, but it seems to be a commonly held view that certain journalists […] are writing under a guide rather than writing off their own bat. Hence the reason they will move the markets – it’s generally fairly easy to tell. (Analyst D)

Another bank analyst dismissed the suggestion that journalists themselves had any direct power but suggested that reporters could inadvertently disseminate market rumours when they called different sources to double-check stories. Nevertheless, he regarded the top analysts as more influential than any financial journalist:

If you think a journalist has power, then the research analyst has incredible power – because their degree of knowledge will be much higher than the financial journalist’s […] so their ability to spread those rumours must be very, very high. (Analyst E)

Another senior bank analyst pointed to a functional symbiotic relation between analysts, reporters, and audiences, but one in which the epistemological power is unequal:

We’re all on first name terms. It ends up being quite a good relationship – they’re trying to gather expert views, expert opinions. But it also means you know who you can trust, and in New Zealand, I have never been misquoted. That’s an interesting misperception, right, that journalists get it wrong and [are] always misquoting people. Well I haven’t, in New Zealand. – Never been misquoted. Not once. (Analyst F)

The fact that this was evidently intended as a complimentary assessment suggests that the role of journalists was regarded as providing accurate stenography in quoting expert sources qualified to define market events. Such attitudes do not indicate a lack of respect for reporters so much as an awareness of how news values shape the possibilities for reporting financial events. Other interviewee observations about the overall quality of financial reporting were mixed. Some interviewees did express concerns about how well journalists could understand financial market complexities, but this was attributed to the small size of the New Zealand market, which limited the scope for the kind of specialist financial reporting possible in the major trading hubs. Although traders and analysts readily acknowledged the importance of (high end) financial media as a source of up-to-date information, this still needed to be interpreted in order to have any investment value, and here the analysts played a key role: ‘Reuters and Bloomberg are experts at […] delivering numbers. But we’re delivering not numbers, but interpretation, which [the media] will follow through’ (Analyst F).

This underlines the point that market reality is not directly accessible or verifiable independently of the market actors whose intersubjective meanings define it. Respondents noted that financially significant information could be filtered out of news reports if it did not fit the genre or format requirements. Consequently, some analysts and institutions deliberately used colourful language to ensure the statements would be quoted or get a headline. Analyst F gave examples of banks putting out statements on the current account deficit using terms like ‘King Kong deficit’ or ‘biggest move in twenty years’, which the news media duly picked up as their headlines, framing the story accordingly: ‘We’re using extreme language and this is a twenty-year event – So they clearly understand; “Wow – this has gotta be on the front of the business page” […] And all the newspapers understood – they knew where to put the story’ (Analyst F).

Another senior bank analyst pointed to the financial media’s need to attribute causes to market events in order to produce coherent stories, even when it is impossible to be certain which factors were driving price movements. He also noted that the angles of stories could evolve throughout the day as more information and points of view were incorporated:

[The media] have to write this much every day – they’ve got to attribute something to something else and you’ll see the stories change as the editions change over the day. The story can change quite dramatically because it’s not actually that easy to find out why [prices] move. (Analyst G)

He also acknowledged that sometimes analysts would suggest explanations of market events to reporters to keep them happy even when this may not reflect their own professional opinions:

You’ve got to give the journalists something they want – you can’t give them something too complicated, too long-winded. They like nice chunks – something that’s really catchy to put in there. Sometimes it doesn’t really capture what they believe, but they know that it’ll get their name out – that’s why they say it.

Trader B reinforced the view that the financial media’s need to define and explain market events can lead to journalists seeking rational explanations for financial events which have no discrete, discernible cause. He also suggested that reporters were under pressure to find stories to report even when nothing of real significance has occurred. Consequently, he sometimes resorted to inventing plausible explanations to satisfy enquiries from persistent journalists:

I’ve had journalists who become extremely annoying – I had a journalist from Dow Jones actually – they would ring me up every single day asking me the same question, say, ‘Why did the market move?’. And then they concoct a story that […] was tenuous at best. And really, sometimes markets don’t move – there’s just nothing going on – And I’d rather they just write there’s nothing going on. […] They may have to release a report at 11 o’clock every day on what’s going on in the market […] so to get them off your back, you’ll feed them some line about, ‘look, this happened’ – but in actual fact you’re not saying it with any great conviction. Bear in mind also that sales teams that are talking to media and customers are going to slant the information in the way they want it to be presented – because they’ve got positions. (Trader B)

Several of the finance professionals interviewed were sceptical about the potential for interactions among analysts, traders, and reporters to directly generate convergences of opinion, herding behaviour, or self-fulfilling prophecies, but such possibilities were not denied outright. Several analysts did acknowledge that information-sharing through their institutional networks intended to find points of consensus and resolve conflicting interpretations might produce feedback loops: ‘There’s quite a close relationship between […] the economists operating together to come up with a consistent view’ (Analyst D). Another bank trader acknowledged that he called analysts at other institutions on a daily basis to check their positions and outlooks (Trader E). Analyst views solicited through professional networks are generally more important than those sourced through the financial news media. However, analyst consensus/average forecasts on key financial indicators such as interest rates are still significant:

Unless there’s a clear trend that [indicates] there’s a herd mentality taking place, then I’ll prefer just to make my own decisions based on my own analysis. Reuters, Dow Jones, and Bloomberg as well, they all release surveys of what the different banks are expecting for interest rate announcements. For every data release they’ll say ‘here’s ten banks, here’s the expectations, this is the average, this is the last number’ – so we all look at those. I don’t need to worry what any one bank thinks – I’d rather just look at the averages. (Trader B)

Interviewees recognised the potential for analysts to influence the market, but were sceptical that any individual would be consistently influential because of the plurality of competing expert claims. There was also recognition that analysts monitored each other quite closely. For example, if a major bank or investment institution changed its forecasts, market reactions would be monitored to see whether other analysts would follow their lead or maintain the previous consensus. This is consistent with Kurtz’s (2000) observations that institutional pressures on analysts tend not to deviate too far from consensus.

The two journalists interviewed were well aware that the financial sources they relied on sometimes had vested interests. Journalist A suggested that information-trading between analysts and journalists meant the relation was, by and large, mutually beneficial. Financial reporters understood that analysts were motivated to build a media profile because there was a ‘competition for ideas’ within the markets and a desire to promote preferred angles or stories. Multiple sources would therefore be cross-referenced before reporting. However, even if multiple sources are consulted, any variations in their views may still conform to the prevailing codes and frameworks within the current market consensus. Interestingly, some traders and analysts cited journalists calling up to check facts as a potential source of market rumours because their questions would often reveal what other market actors were thinking. Another journalist suggested that cultivating personal relationships with sources was desirable for building trust and gaining exclusive stories but noted this was often not possible since ‘we all talk to the same people’ (Journalist B). The journalist went on to suggest that, ‘you can tell when they’re talking their book – you’ve got to be a bit cynical’.

However, it was acknowledged that the expertise of sources was often superior to the reporters’ and that some sources were intolerant of critical interrogation of their claims. Indeed, a key institutional primary definer mentioned was the ratings agencies (notably Standard and Poor’s, Moody’s, and Fitch), which journalist B acknowledged were often regarded as ‘God’. This would certainly suggest a limited scope for reporting financial events from perspectives that do not align with the epistemological premises of finance professionals. The journalists’ observations also underline the point that financial reality is not directly accessible and requires articulation by primary definers within the financial system and, moreover, that verification is often based on corroboration by other experts. This does not preclude publication of news reports critical of the financial sector, but it does suggest that financial news as a genre is expected to reflect the codes and schemata of financial primary definers in order to be deemed salient and meaningful. In this regard, in line with Parsons (1989), financial news discourses play a role in reinforcing and legitimating the prevailing frameworks of knowledge among market actors.

Reflexivity, representation and reporting – some conclusions

The preceding discussion has demonstrated the importance of recognising the reflexive, constitutive role of information in financial market processes. Coupled with the analysis of the institutionalised relationships among financial reporters, analysts, and traders, it becomes apparent that the neoclassical model of informational efficiency and positivist notions of objectivity in news representation are not sustainable. In particular, the process of verification is problematic in respect of financial events which are not accessible independently of the epistemologies and technologies of appresentation which underpin their intersubjective validity.

The financial media are structurally predisposed to reinforce market consensus by focusing market attention on particular stories or frames and providing the context for interpreting financial news. This does not mean that the financial media uncritically reproduce the specific views of analyst sources which then trigger trading responses. Nor does the relation among traders, analysts, and journalists generate simple self-referential flows of information that directly generate herding behaviour and cause crises. However, it is important to recognise that the ostensible plurality of source opinions solicited by financial journalists may still conform to the prevailing epistemology underpinning current trading frames or schemata. It is on this level that the financial media are most likely to be complicit in the reinforcement of market consensus and contribute to the formation of system-level feedback loops which, as Minsky (1977) recognised, play a role in the cyclical formation of bubbles and crashes. Specifically, the reinforcement of the prevailing consensus as fictitious asset values are expanding, promotes investment frames that interpret the expansion as confirmation of the prevailing valuation models and discount the risk of increasing fragility.

Financial reporting faces its most complex challenge at the transition points in market cycles or during periods of instability, because this is where the intersubjective validity of the prevailing trading frames or schemata can break down. This means the potential for investor uncertainty to either stabilise or exacerbate into a crisis cannot be verified independently of market actors’ collective responses to unfolding events. During such periods, it makes sense for investors to monitor other market actors in order to discern any shift in consensus expectations and the variables driving trading, because the continuing validity of valuation and trading models or schemata depends on aggregate market reactions. In the recent crisis, for example, the ratings agencies’ downgrading of many collateralized mortgage securities from investment grade to junk resulted in a break-down of the intersubjective codes which permitted their coherent valuation (Thompson, 2010b).

Meanwhile, real-time information flows and market responses compress the time-frames available for information processing and decision-making both for investors and policy-makers, as well as intensifying the financial media’s prioritisation of ongoing updates on unfolding events (which potentially explains the displacement of more critical, contextual ‘big picture’ analyses). Even though professional investors and policy-makers do not generally rely on public news media as the basis for trading or policy decisions, their perceptions of evolving market sentiment are likely to be contextualised by the way events are framed in the news (see Bryan, 2001). Consequently, the legitimation and efficacy of the massive government bail-outs intended to restabilise the credit system (such as the Troubled Asset Relief Program), depended partly on the investment community’s collective confidence in such measures which in turn depended on the shifting direction of market consensus. Clearly, the financial media cannot be held responsible either for the crisis or for the subsequent government responses. Nevertheless, the tendency to reflect and reinforce the consensus views of elite financial sources suggests that, just when the demand for accurate and reliable financial news is most acute, the possibility of providing it is complicated by the processes of informational reflexivity.