https://orcid.org/0000-0001-5481-8906
![Sample our Economics, Finance,Business & Industry journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5931/TOC-Subject-Sample-2021-Economics-Finance-Business-Industry.jpg"})
Abstract
Weber famously invoked ‘ideal types’ as an analytic device with which to measure empirical reality against some hyper-rational fabrication. Case in point: non-professional (lay) investors appear to be the antithesis of rational economic man. They have been cast as less-informed, less-skilled, and less-knowledgeable than professional market practitioners, and with ample evidence that they tend to lose money in the market as a result. This study builds the case that a new class of algorithmic financial advisor, commonly known as ‘roboadvisors’, enacts lay investors as rational market actors. This is achieved through algorithmic devotion to modern portfolio theory (MPT), which the roboadvisors embody, automate and perform, conjuring some version of homo economicus into existence. Through this example, I show how Weberian ideal types and the particular kind of rational action associated with them (e.g. the ideal type investor) become the very empirical reality they were intended to be a foil to – accomplished through the technological articulation of financial models, even in the hands of ordinary individuals.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Adam Hayes is a Ph.D. candidate in Sociology at the University of Wisconsin–Madison. His research investigates sociological factors that influence financial decision-making and structure economic behaviour that have previously gone unnoticed. Adam’s scholarly work falls at the intersection of economic sociology, the social studies of finance and behavioural economics, with recent publications based on this work appearing in Social Forces, Socio-Economic Review, Sociological Theory and Theory, Culture, and Society. In addition to his current graduate work, Adam also holds an MA in economics and is a CFA (chartered financial analyst) charterholder.
Notes
1 Weber initially conceived of ‘ideal types’ in his 1904 essay on Objectivity which has been identified by scholars to refer specifically to his theory of history (e.g. Kalberg, Citation1994). The Economy and Society version of ideal types, on the other hand, was fashioned exclusively with sociology in mind (see: Swedberg, Citation2018, p. 182; Schutz, Citation1967; Hekman, Citation1983). In this paper, I consider exclusively the ‘sociological’ rendition of ideal types.
2 Weber indeed identifies several dozen ideal types in his writings, including the ideal type bureaucracy and ideal type capitalist entrepreneur, but devotes most space to economic action.
3 Weber (Citation1978[Citation1922], p. 26) instructs that, ‘action is instrumentally rational (zweckrational) when the end, the means, and the secondary results are all rationally taken into account and weighed. This involves rational consideration of alternative means to the end, of the relations of the end to the secondary consequences, and finally of the relative importance of different possible ends’. Swedberg (Citation2018) adds that an ideal type actor has full knowledge of the situation and is fully aware of what he/she is doing.
4 This point is elaborated in greater detail in the discussion that follows the empirical section.
5 In fact, in the case of MPT asset price does not even come into play, only portfolio weights.
6 While this is true of the sociological version of Weber’s ideal type, the historical version has been more frequently employed, especially in comparative-historical work (e.g. Kalberg, Citation1994; Biernacki, Citation2012; see also endnote 1).
7 Weber acknowledges that ideal types can alternatively be oriented to other modes of action (specifically: value-rational; affectual; and traditional), but that to be an ideal type only one mode of meaningful action is chosen. It is by identifying other modes of meaning in contrast to the ideal type that sociological inquiry can proceed.
8 In practice, MPT would allocate dollars to representative indices (i.e. ETFs or mutual funds) corresponding to a specific asset class, including (e.g.): domestic/foreign/emerging markets large-/mid-/small-/micro-cap stocks; domestic/foreign corporate bonds (of various credit rating); domestic/foreign government debt; commodities (e.g. gold, silver, oil, etc.); domestic/foreign real estate (e.g. REITs); and so on.
For a detailed account of the development of Markowitz's work on MPT, see chapter 2 of Bernstein, Citation1992.
9 Efficient frontier data points obtained from www.portfoliovisualizer.com
13 Since roboadvisors manage client money for lay investors, they fall under the regulatory scrutiny of the securities and exchange commission (SEC) as registered financial advisors. Because of this they are required to file mandatory disclosures speaking to the investment methodology used and the extent to which algorithms are deployed in their practice, in what is known as the Form ADV Part 2 Brochure.
15 Personal Capital requires a starting balance of $100,000, Vanguard Personal Advisor $50,000, Zack’s Advantage because it uses Schwab’s roboadvisor to implement its allocation, where I already had an account established, and Honest Dollar which only offers retirement accounts.
16 Operationalized as annualized standard deviation, or the square root of the variance over the same period.
Roboadvisors are a relatively new phenomenon with the majority of those in my sample launching since the year 2014. Therefore, this study period was used to capture overlapping multi-year data covering my entire sample.
17 Portfolio returns are pre-tax annualized historical returns from 1 January 2015 through 31 December 2017. Portfolio standard deviations of returns are annualized over the same period.
19 Since lazy portfolios are low-maintenance set-it-and-forget it allocations there is no meaningful annual cost associated with them.
21 These data for the study period were sourced from Research Affiliates, LLC’s (RA) Asset Allocation Interactive Tool. RA is an investment manager and financial analytics firm that provides its services mainly to investment companies. See also Shepherd et al. (Citation2018). https://interactive.researchaffiliates.com/asset-allocation#!/?category=Model¤cy=USD&model=ER&scale=LINEAR&selected=225&terms=REAL&type=Portfolios
22 It is revealing that having a human advisor produces, on average, the least efficient portfolios, especially on an after-fee basis.
23 This is an important distinction that sets the like of roboadvisors apart from other passive indexed strategies such as target-date funds, which take a one-size-fits-all approach. Thus, while a particular target-date fund that matures in the year 2035 may prove optimal for some small subset of investors, its underlying portfolio will fail to be the rational (efficient) choice for many others – even if they do plan to retire in that year.
24 In the most basic form, a buyer of securities must confront a seller and vice-versa.
25 From my own engagement with these platforms, this choice seems to boil down essentially to cost, since all the roboadvisors are effectively doing the same thing (see ).
26 My gratitude to the anonymous reviewer who pointed out this apparent paradox.
27 It is worth noting that just as too much MPT investing can lead to instability, so too can a large proportion of high-frequency trading – which has been blamed for the occurrence of ‘flash-crashes’ (Lewis, Citation2015; Lange et al., Citation2018), sudden price drops facilitated by negative feedback loops where certain HFT algorithms trigger others in a downward spiral. But, interestingly, it may be the HFT’s active orientation that balances the potential dangers of MPT-following algorithms. It is important to note that HFTs already trade against roboadvisor order flow since the former acts as market-maker in many ETFs. Ultimately, a new market logic may establish itself where the more passive algorithms of roboadvisors and the active ones of high-frequency trading end up supporting and sustaining one another, as well as promoting systemic stability as they trade against each other.
28 These examples are similar to ‘paradoxes of rationality’ invoked in game theory; see e.g. Grüne-Yanoff, Citation2012.
29 ESG stands for ‘environmental’, ‘social’ and ‘governance’.