Introduction
As of late, the terms ‘flexible work’ and ‘gig economy’ have acquired at least as many negative connotations as positive ones.Footnote 1 While some intuitively associate these notions with organisational innovation, individual empowerment and flexible work-life balance,Footnote 2 for others they are emblematic of rampant worker exploitation, individual vulnerability and persistent economic insecurity.Footnote 3 Although the main forms of flexible work practice observed in today's economy are by no means novel phenomena, both casual and gig-based work arrangements have undoubtedly been subject to unprecedented levels of attention recently.
Arguably the principal reason for the increased attention devoted to such practices today has been the widely-reported usage by many well-known corporate employers of so-called ‘zero-hours’ contracts, which are commonly perceived as denying workers any basic level of security with respect to employment tenure or continuing receipt of work income.Footnote 4 The enhanced public salience of flexible work practices is additionally attributable to high-profile UK employment tribunal and court cases on alleged employer exploitation of gig economy workers. Indeed, the question of whether gig economy operatives should be classified, for employment protection purposes, as either workers or independent contractors has proved to be one of the most vexing legal and public policy concerns of recent times, and has prompted a rapidly growing line of judicial decisions on the question of bogus self-employment.Footnote 5 The above issues were also the fulcrum of the government-commissioned Taylor Review of Modern Working PracticesFootnote 6 in 2017 and the subsequent report of the House of Commons Work and Pensions and Business, Energy and Industrial Strategy CommitteesFootnote 7 into the same matter, which presently looks set to prompt significant reform to key aspects of UK employment law.Footnote 8
It is not the purpose of this paper to undertake any sort of detailed technical analysis of these legal and regulatory developments, which have already been (and no doubt will continue to be) subject to rigorous critique elsewhere.Footnote 9 Above all, the evolving approach of tribunals and courts to determining the presence of employer direction and control over app-based gig workers, along with the implications of the Taylor Review's proposed new ‘dependent contractor’ status,Footnote 10 are likely to command the considerable and sustained attention of employment lawyers over the coming months and even years. In acknowledgement of this fact, and with a view to further enriching the conceptual soil for such debates, the present paper deliberately seeks to chart an alternative methodological course with respect to the topic at hand, one which is probably more redolent of established corporate rather than employment law scholarship.
Drawing on key concepts from law and economics literature and, in particular, Ronald Coase's classic work on the economic nature of the business firm, this paper seeks to position the main forms of flexible work within a broader institutional frame of reference. Accordingly, it focuses principally on the implications of task-oriented and on-demand work in terms of the relative allocation of control, risk and reward within the firm. From this perspective it will highlight how both task-oriented and on-demand work arrangements typically entail workers assuming a significant degree of residual economic exposure to product market uncertainty and other exogenous shocks, absent any corresponding compensation in the form of enhanced contractual, proprietary or governance entitlements.Footnote 11 From an employer point of view, flexible (and especially zero-hours) work is thus a hybrid phenomenon that combines the most favourable features of both firm and market-centred forms of human capital allocation. However, from a worker perspective the opposite will frequently be the case.
Adopting a hypothetical contract analysis (likewise inspired by Coase's work), the paper will subsequently enquire as to optimal forms of quasi-contractual compensation that rational flexible workers can (counter-factually) be regarded as bargaining for in the absence of material impediments to efficient contracting, and as the notional price for trading off their traditional employment guarantees. The aim of this enquiry is to provide fresh normative insights into what is already a highly fertile field for legal-academic analysis, with a view to enhancing the inter-disciplinarity and conceptual sophistication of future scholarly debates and regulatory reform initiatives. The discussion accordingly proceeds as follows.
The paper begins by highlighting the key characteristics of the two orthodox forms of productive human capital allocation from a Coasean firm-theory perspective, namely employment and self-employment. However, whereas Coase analysed the distributive implications of the employment relation more or less exclusively from the human capital user (ie employer) side of the picture, this paper focuses instead on the distributive consequences of employment (vis-à-vis self-employment) for the relevant human capital supplier (ie employee). Next, it extends this analysis by reference to two of the principal forms of human capital provision that lie between these extremes, namely task-oriented (gig) and on-demand (casual) work, identifying their respective similarities to, and also differences from, employment and self-employment in the traditional sense. On this basis, it then develops a hypothetical bargaining model geared to determining the notionally ideal means of compensating flexible workers for their assumption of residual business risk exposure and associated positional costs, in particular by drawing a structural comparison between, on the one hand, key forms of flexible or non-standard work, and, on the other, ‘at-risk’ or unsecured financial investments in business firms. Finally, it provides some concluding observations.
1. Employment versus self-employment as competing forms of productive human capital allocation
In his classic 1937 article ‘The nature of the firm’,Footnote 12 the economist Ronald CoaseFootnote 13 contrasted the key characteristics of markets and firms as the basic institutional means of allocating human capital within productive activity.Footnote 14 Accordingly, Coase reasoned that an entrepreneur, in determining how to obtain the necessary factors of production for her business, is in the simplest case faced with a choice between two alternatives.Footnote 15
One option is for the entrepreneur to conduct her business purely by way of the open market, obtaining each factor of production that she requires by way of ad hoc negotiation with various suppliers. If the entrepreneur is dependent exclusively on physical capital (eg premises, equipment and/or stock in trade) to sustain her business operations, and is both able and willing to supply all of the necessary human capital (ie labour or manpower) herself, then such a strategy makes perfect sense. But if the entrepreneur is reliant to a significant extent on the labour of others for her productive activities, then a purely market centred approach to hiring human capital becomes much more difficult to sustain.
In theory the entrepreneur could conceivably still opt to obtain all of the human capital she needs on a decentralised market basis, by negotiating a discrete (or ‘spot’) supply contract every time she requires a particular productive, logistical or administrative task fulfilled within her business (or at least one that she is unable or unwilling to carry out herself). Almost certainly the more rational option in this scenario, though, would be for her simply to enter into a few longer-term and relatively indiscrete (ie open-ended) contracts with certain trusted human capital suppliers, under which they each agree to submit to the authority and direction of the entrepreneur within contractually agreed bounds.Footnote 16
From the point of view of allocating human capital in the service of productive activity, the institution known as ‘employment’ thus essentially entails supplanting the price mechanism of the outside labour market with an alternative hierarchical coordination mechanism known as ‘the firm’.Footnote 17 Accordingly, within the ambit of the employment relation the input-provider (as employee) becomes subject to the discretionary prerogative of her employer on a continuing basis.Footnote 18 This obviates the considerable transaction costs her employer would otherwise have to incur to obtain the same degree of labour by way of a series of individual task-based contracts with outside human capital suppliers.Footnote 19
Such may be the case from the human capital hirer (or employer)’s point of view, but what about from the perspective of the human capital supplier? Curiously, Coase seemed to devote less direct consideration to the latter issue. Notwithstanding, the essential nature of the corresponding dilemma confronting the latter party to the relation can readily be surmised as follows.
On the one hand the human capital supplier might opt to set up as an entrepreneur or firm owner herself, thereby providing her labour (or, more accurately, the goods or services produced or supplied via her labour) directly on the open market and independently of any superior employer coordination or oversight. Accordingly, the human capital supplier (as entrepreneur) will purchase or hire the necessary physical capital and ancillary human capital (eg logistical or administrative work) to support her in her productive activities, while retaining the ultimate power of direction over her work along with the associated ownership right to appropriate the residual profit of her business.Footnote 20 At the same time she will have the unbridled right to diversify her human capital as widely or as narrowly as she chooses, subject only to the obvious extraneous limitations of physical and market capacity. Quid pro quo, the human capital supplier (as entrepreneur) will undertake to bear the ensuing residual risk of loss in the event that her business venture either fails or is lossmaking, including the inevitable depreciation and replacement costs incurred in connection with any physical capital which she owns. However, this will be counterbalanced by her corresponding capacity – as residual firm controller – to wield a determinative influence over the firm's business strategies and, in particular, its prevailing disposition towards risk taking.Footnote 21 On the other hand, the human capital supplier (as entrepreneur) will have no ex ante contractual guarantee of paid work in the vein of an orthodox employee, but rather will be compelled to seek out prospective ‘purchasers’ of her labour directly on the open market while absorbing the ongoing search, negotiation and other costs that this entails. Alternatively, the same human capital supplier might well decide (assuming that the opportunity is available to her) in effect to opt out of the open market altogether, by providing her labour instead within the context of a unitary employment (or firm) relation. The upsides to her of doing so are readily apparent. In the orthodox employment relation, the human capital supplier (as employee) is promised a pre-agreed rate of remuneration by her employer for the duration of the working relationship, and irrespective of interim changes in the relevant market environment.Footnote 22 The employee consequently enjoys a degree of relative immunity from residual business risk, insofar as her rate of return on human capital is – at least in the immediate term – unaffected by adverse product market and other extraneous developments.Footnote 23 Finally, the employee enjoys additional immunity from human capital-redeployment costs insofar as her relative security of income obviates the need to seek alternative buyers for her labour (or, rather, the product(s) thereof) on the outside market for services.
Of course these valuable immunities do not come for free,Footnote 24 but rather are gained at the corresponding cost (for the human capital supplier) of having to submit her labour to the ultimate direction and oversight of an employer.Footnote 25 This typically entails the employee surrendering determinative control over not just what she does ‘on the job’, in the sense of what specific work tasks she performs and how she carries them out.Footnote 26 It also tends to involve the employee ceding a significant degree of her ‘off the job’ freedom, to the extent that her obligation to be available during specified working hours has the effect of restricting her practical capacity for diversification of human capital across different and unrelated work outlets. Furthermore, with her relative immunity from residual business risk exposure comes the employee's corresponding exclusion from the management and governance of the firm, such that the employee (at least qua employee) is ordinarily denied the right to influence the strategic direction of her employer's business and – relatedly – its prevailing risk appetite.Footnote 27
In Coasean terms, therefore, the transition from human capital entrepreneurship to employment entails the relevant capital provider effectively ‘trading off’ her rights to control and diversification of human capital, in return for relative insulation from residual business risk exposure and capital-redeployment costs. Moreover, as the employee (unlike the human capital entrepreneur) is no longer the ultimate owner of the business in which her labour is employed, it goes without saying that she ordinarily enjoys no converse upside right either to appropriate any share of residual business profit, or to influence the overall strategic direction and risk appetite of the firm's business. The essential distinctions between these two typical economic role definitions can accordingly be summarised in the following basic typology.
Table 1: Key structural features of self-employment versus orthodox employment
Note: 1 denotes typically strong possession of the relevant attribute; 0 denotes typically weak or non-possession
2. Task-oriented and on-demand work as hybridised firm/market phenomena
The Coasean distinction between employment and self-employment as competing forms of productive human capital allocation is highly instructive. However, at least in its basic form the model is ultimately one of extreme types; in reality, though, the variation between the two phenomena is typically a sliding scale of multiple hybrid forms, as opposed to a stark binary dichotomy.Footnote 28 Meanwhile, the economic actors typically employed in such intermediate roles today can vary from the one extreme of the occasional student gig worker seeking additional income for leisure activities, to the sole-occupation zero-hours contractor looking for an adequate level of monthly income to house, feed and/or clothe herself and her dependents.Footnote 29
However, a uniting feature of all such atypical work arrangements, notwithstanding their significant differences, is the fact that they commonly exhibit some combination or other of certain elements of employment and self-employment in the traditional sense of the terms. As such they are aptly characterised as hybridFootnote 30 (or, perhaps more appropriately given the usual bargaining dynamics involved, hybridised) firm/market phenomena.Footnote 31 Moreover, broadly speaking it is possible to delineate two particularly common hybridised quasi-employment forms in this regard, namely: (i) task-oriented (or gig) work, and (ii) on-demand (or casual) word. As will be explained further below, whereas the former type of work tends to exhibit many (albeit by no means all) characteristic features of market-centred human capital allocation (ie self-employment) in the Coasean sense, the latter is typically much more akin to standard firm-centred human capital allocation (ie employment), at least insofar as its negative (or ‘downside’) payoffs for the relevant worker are concerned.
(a) Task-oriented (gig) work
The term ‘gig economy’ is often used today in a broad sense to denote what is widely perceived as a contemporary labour market climate dominated in large part by short-term and insecure working arrangements.Footnote 32 However, this is something of a misnomer. Strictly speaking, the notion of ‘gig’ work refers specifically to task-oriented or ‘piece’ work:Footnote 33 that is, where a worker receives payment for the performance of discrete and divisible single functions such as individual taxi journeys, goods deliveries or domestic tasks.Footnote 34 This is in contrast to the scenario where the worker subjects herself generally to the directions or orders of an employer over an agreed time period, as in the case of traditional wage labour.Footnote 35
The term is also commonly used in a more specific sense today to denote the purchase and sale of divisible labour online by way of smartphone and tablet apps,Footnote 36 such that it is tempting to see gig work as being one of the peculiar social products of the modern digital age. But despite its vogue present-day connotations, the notion of the gig economy – in the sense of a productive paradigm characterised to a large extent by granular, task-oriented work – is by no means a novel phenomenon.Footnote 37 In fact, far from it, the notion long predates the dawn of industrial capitalism itself, and has its roots in the elementary homeworking practices of the pre-Victorian craft and textile trades.Footnote 38
Although task-oriented or ‘gig’ work in the above sense takes many and varied forms today, the most well-documented examples of such arrangements tend to share some common structural characteristics. These features, moreover, are remarkable when examined from the perspective of the Coasean market versus firm (or self-employment versus orthodox employment) dichotomy. One of the most noteworthy characteristics of task-oriented work, which is intrinsic to its very nature, is that it entails no obligation on the part of the employer to ensure the continuing availability of paid work for the duration of the work relation. Rather, ongoing work-availability (from the worker's perspective) is dependent entirely on the prevailing level of demand for their labour within the relevant geographic or other spatial locality.
The worker's ability to sustain their anticipated rate of return on human capital is consequently rendered directly contingent on preservation of existing product market conditions, meaning that they are continuously exposed to the residual risk of future adverse change in this regard (eg as a result of increased supply of the same labour and/or developments precipitating reduced market demand for the relevant service).Footnote 39 Moreover, insofar as gig workers are customarily expected to commit their own physical (as well as human) capital to the relevant enterprise (eg a private car, van or bicycle in the case of a gig-based taxi or delivery firm, or self-purchased tools or other equipment in the case of a gig-based service firm),Footnote 40 they are further exposed to the depreciation and replacement costs associated with personal ownership of those assets.Footnote 41 By contrast, the orthodox employee enjoys relative buffering from residual business risk via a combination of: (i) the continuing contractual guarantee of a certain level of paid work for the duration of the employment relation, irrespective of interim fluctuations in prevailing market equilibria; and also (ii) her employer's ownership of the principal physical capital deployed in the business, together with the accompanying exposure to asset depreciation and replacement costs.Footnote 42
The task-oriented worker is further compelled to absorb ongoing human capital redeployment costs to the extent that any anticipated shortfall in labour income must be offset by additional income from other sources of work. It follows that, at least insofar as her exposure to residual business risk and human capital redeployment costs is concerned; the task-oriented worker is much more akin to a human capital entrepreneurFootnote 43 than the orthodox employee of the Coasean model. As regards compensation for her labour, the task-oriented worker is customarily entitled to a price for each individual task performed, with the applicable rate of pay determined unilaterally by the employer. While this enables the worker (like an entrepreneur) to appropriate a level of income in direct proportion to the extent of her committed labour, the employer nonetheless typically retains the capacity to appropriate surplus firm value over and above the worker's pre-determined share.Footnote 44 The worker is consequently denied the right to appropriate residual business profit as reciprocal ‘upside’ compensation for absorbing its ‘downside’ residual risk.Footnote 45
Additionally, unlike the human capital entrepreneur of the Coasean model, the task-oriented worker normally has no right to influence the running of the overall business in which her labour is employed. Insofar as this precludes the worker from influencing the firm's prevailing risk appetite, it means that the worker's continuing risk-adjusted rate of return on human capital is (indirectly) subject to variation at the unilateral prerogative of the employer. In economic terms this represents an uncompensated social cost that the task-oriented worker is compelled to bear, which – moreover – is impossible for her to price ex ante (eg via a correspondingly higher agreed piece rate for her work) due to the inherent uncertainty and indeterminacy of the various market dynamics and other factors at play.Footnote 46
Moreover, in numerous reported instances it would appear that task-oriented workers do not even have meaningful control over the allocation of their human capital itself, let alone the direction of the overall business enterprise in which it is employed. Indeed, as controversial recent cases on the legal classification of gig workers have demonstrated, many gig workers are still subject to significant employee direction and/or control regarding specifically how, and how frequently, their work tasks are performed. The effect of such measures is to constrain the task-oriented worker's freedom with respect to ongoing allocation of her human capital (at least while ‘on the job’), such that one of the traditional positive payoffs from assuming self-employed status becomes largely negated.Footnote 47
On the other hand, in theory the principal positive payoff for workers of assuming task-oriented work (at least relative to orthodox employment) is the degree of ‘off-the-job’ freedom that it purportedly permits in the sense of enabling gig workers to acquire a varied ‘portfolio’ of income-generating work sources, thereby providing scope for diversification of their labour and consequent ‘hedging’ against firm-specific risk.Footnote 48 However, while the notion of gig workers as ‘portfolio’ human capital investorsFootnote 49 might have a certain intuitive theoretical appeal, the realities of task-oriented work tend to paint a much bleaker picture in this regard. It has been found that the practical capacity of task-oriented workers to provide their services via multiple work platforms is dependent ultimately on the nature of the work that they are equipped to perform. While certain types of gig work such as virtual PA and IT-based administrative work are in general more susceptible to being part of a varied multi-job portfolio, this tends not to be the case for more geographically restricted forms of gig work and/or those that necessarily entail a direct physical interface with customers, such as taxi-driving and courier work. Accordingly, those workers operating in the latter types of sector have tended to find themselves more exposed to the risk of adverse fluctuations in local labour (and, concomitantly, product) market conditions.Footnote 50 Moreover, recent evidence (from 2017) has shown that, in the UK at least, the phenomenon of the diversified portfolio gig worker is by and large a fictional creation, with over 60% of UK-based gig workers (including three in four female gig workers) earning less than £11,500 per year from task-oriented work, and over 30% earning less than £4,500 per year from such work.Footnote 51 Meanwhile, the findings of a recent (2018) government survey have revealed, inter alia, that the median annual income of gig economy workers in the UK is only £375, with 65% of respondents reported to have earned less than 5% of their total income over the past year from gig work.Footnote 52
As a result, over 60% of UK-based gig workers – including 73% of gig workers aged between 31 and 54 years old – have been compelled to rely on more conventional forms of work to supplement their limited income derived from task-oriented work alone. Indeed, a recent survey has revealed that the vast majority of gig workers in the UK work only 16 hours per week or less in this way, with half accepting less than one paid gig per month.Footnote 53 The implication is that, at least in its present form and scale, the task-oriented labour market is largely incapable of providing gig workers with meaningful scope for inter-firm diversification of their human capital, thereby significantly restricting their capacity for effective risk-hedging at the individual firm level.Footnote 54
(b) On-demand (casual) work
So-called ‘casual’ or on-demand work arrangements, while often viewed as analogous to gig work in the above sense, are really a qualitatively different phenomenon altogether. In fact, as will be demonstrated below, in their most extreme and well-known form today, namely so-called ‘zero-hours’ contracts, they are more akin to standard employment relations than task-oriented work. As in the case of task-oriented work, the time periods over which on-demand (and especially zero-hours) work will be made available tend not to be contractually specified in advance. Rather, as its name suggests, on-demand work is customarily offered on a purely ad hoc basis at the short-term (in many instances ‘on-the-day’) request of the employer. However, the key difference from task-oriented work is that, with on-demand work, the worker's person itself is typically regarded as being committed to the work. Accordingly, the on-demand worker (like an orthodox employee) is usually paid for their labour generally as opposed to any specific discernible ‘output’ that their work produces.
The typical task-diffuseness of on-demand work (at least relative to task-oriented work) is attributable in large part to the inherent nature of the work itself, which – unlike many types of gig workFootnote 55 – tends to be geographically-specific in the sense of needing to be performed in a specific physical location such as a shop, factory or warehouse.Footnote 56 On-demand work is also, to varying extents, typically less remote or autonomous than task-oriented work in the sense that many common types of on-demand work – such as retail, customer service or storage work – are dependent on the coordinated interaction of cohesive groups of workers, whose collective marginal product is consequently worth more than the simple aggregate of each of their respective individual inputs (to the extent that the latter is even capable of reliable calculation).
Therefore although on-demand work has – like task-oriented work – been described as ‘divisible’ in nature,Footnote 57 it is divisible in a somewhat different sense. Whereas the latter type of work (as explained above) is divisible at the micro level of the prescribed worker task or output itself, the former kind of work is divisible in the slightly different sense that on-demand workers are compensated on an impromptu periodic (eg hourly or daily) basis rather than by way of integral compensation for their continuing labour generally (eg by way of a fixed annual salary). In both cases, therefore, the relevant work arrangement (on the supply side at least) exhibits a significantly greater degree of ‘granularity’ (or ‘bittiness’) than the orthodox employment relation as depicted in the Coasean model.Footnote 58
In terms of the particular bundle of structural features that it typically exhibits, on-demand work tends to represent an even more lopsided apportionment (in the employer's favour) of the social costs and benefits of productive activity than task-oriented work. As in the case of task-oriented work, with on-demand work there is no employer obligation to provide a minimum level of continuing paid work throughout the duration of the work relation (hence the abovementioned use of the popular term ‘zero-hours’ to refer to the most extreme forms of such work arrangementFootnote 59). Furthermore, whereas the task-oriented worker tends at least to have a reasonable degree of ‘off the job’ freedom with respect to when and also how frequently work tasks are performed, this is ordinarily not the case for the on-demand worker. Rather, the fact that the on-demand worker is generally expected (albeit often not formally required as suchFootnote 60) to be available for work during requested time periodsFootnote 61 significantly constrains the degree of control they have over the allocation of their human capital, both within the relevant working relationship and – moreover – outside of it.Footnote 62 It follows that the on-demand worker is typically much more restricted than the task-oriented worker as regards her ongoing availability for additional paid work from other sources.Footnote 63
From the perspective of the present analysis, a notable consequence of the above geographic and institutional constraints is that on-demand work is, at least in the standard case, necessarily even more firm-specific in nature than task-oriented work due to the practical incapacity of the typical on-demand worker to reallocate her labour to alternative productive outlets (at least without incurring significant displacement costs in doing so). This ‘lock-in’ propensity of on-demand work has the effect of significantly constraining (if not eradicating outright) the on-demand worker's practical capacity for human capital diversification and consequent risk-hedging, typically to an even greater extent than her task-oriented counterpart.Footnote 64
Generally speaking, on-demand work is also even less advantageous for workers than task-oriented work when it comes to distributing the upside and downside aspects of residual business risk exposure. Like the task-oriented worker, the on-demand worker's anticipated rate of return on human capital is subject to continual variation in accordance with the prevailing level of demand for their labour. Insofar as demand for their labour at any one time is dependent – in turn – on corresponding fluctuations in relevant product market demand, it follows that the on-demand worker is likewise exposed directly to the residual business risk of their employer.Footnote 65 Such residual risk exposure is especially pertinent in the case of zero-hours workers, who by definition lack any downside contractual insurance against intermittent market shocks. However, whereas the task-oriented worker at least enjoys the limited capacity (subject to continuing demand for the product(s) of her labour) to enhance her income via voluntary assumption of increased tasks, in the standard case of on-demand (and particularly zero-hours) work, the prerogative to determine a worker's current time allocation rests unilaterally with her employer. Furthermore, insofar as adverse variations in worker time allocation risk occasioning frequent material shortfalls in work income, the on-demand worker is similarly exposed to continuing human capital-redeployment costs in the form of the need to secure additional sources of labour, albeit (for reasons explained above) with a likely lesser degree of flexibility in this regard than her task-oriented counterpart.
Finally, it perhaps goes without saying that the on-demand worker ordinarily has no entitlement to influence either the general strategic direction of her employer's business, or its assumed level of enterprise risk exposure, via involvement in management or governance of the firm. This is in spite of the fact that, as a substantially open-ended contractor with no ex ante immunity against residual business risk exposure, the on-demand worker's risk-adjusted rate of return on human capital is (like that of her task-oriented counterpart) susceptible to unilateral variation in accordance with the prevailing risk appetite of their employer's management. Moreover, the on-demand worker's relatively limited capacity for diversification of her human capital (as remarked on above), for instance in less risky work outlets, arguably heightens her unilateral contractual vulnerability along with the consequent need for effective surrogate measures to compensate her in this regard.
In summary, then, the essential features of (i) task-oriented and (ii) on-demand work, when analysed from a Coasean firm-theory perspective, are set out – respectively – as follows (for purposes of comparison, the corresponding respective features of self-employment and orthodox employment are additionally restated below).
Table 2: Key structural features of self-employment versus task-oriented work versus on-demand work versus orthodox employment
Note: 1 denotes typically strong possession of the relevant attribute; 0.5 denotes typically variable or inconsistent possession; 0 denotes typically weak or non-possession
3. Compensating flexible labour: a hypothetical bargain perspective
Although the above binary typology is intentionally crude and generalised, it is nonetheless effective in highlighting certain curious features of those hybridised work forms that are most commonly observed in practice. In short, it would appear that almost all the key disadvantageous (ie ‘0’) features of self-employment and orthodox employment from a worker perspective are substantially replicated within the two main hybridised forms of those relations (that is, task-oriented and on-demand work). However, by contrast, it would seem that – with only very limited exception – the main advantageous (ie ‘1’) aspects of self-employment and orthodox employment (for workers) do not tend to be replicated within the hybridised forms. The normative implication is that these hybridised forms are systematically sub-optimal, in the sense that the ensuing negative trade-offs for workers are rarely (if ever) compensated for effectively by way of corresponding contractual or other relational benefits.
Of course, in the (assumed) absence of material power imbalance and other impediments to efficient contracting by the parties concerned, such a skewed distribution of costs and benefits between employers and hybridised workers need not in itself constitute a socially sub-optimal outcome. Under such (idealised) bargaining conditions,Footnote 66 the contracting parties would presumably be able to both ‘price’ and, if desired, mutually ‘trade off’ their respective entitlements in reaching a distributive outcome that maximises the joint surplus accruing from their productive activities. As a result, the initial (ie legal or pre-contractual) allocation of entitlements between them should be unimportant.Footnote 67
Accordingly, if – to adopt a hypothetical example – subjecting a worker to pervasive employer direction and/or control is conducive to minimising her employer's costs of productionFootnote 68 (eg by enabling quicker detection by the employer of reduced effort or ‘shirking’Footnote 69 on the worker's part), the employer will rationally seek to impose such an arrangement on the worker concerned.Footnote 70 However the worker, far from being oppressed or exploited by this arrangement, will rather only rationally submit to the employer's terms if adequately compensated for doing so. For instance the worker might agree to accept reduced ‘on the job’ freedom in return for a correspondingly higher rate of pay from her employer. Moreover, such a voluntary trade-off is possible irrespective of the parties’ initial legal position, which (at least in theory) can readily be varied in accordance with their rational individual preferences.Footnote 71
It follows that, so long as the anticipated cost-savings for the employer (from reduced worker shirking) exceed the compensatory wage increase demanded by the worker, the arrangement will produce a net gain for the employer. And, likewise, insofar as the worker values her promised wage increase more than the ‘on the job’ freedom she would otherwise have enjoyed, the same arrangement is also conducive to a net gain for the worker. In this way, the contracting parties’ respective preferences are brought into equilibrium in a manner that maximises their joint economic surplus, in the sense of the combined net value of their productive endeavour.Footnote 72
By the same token, requiring the notional worker to bear a significant degree of residual business risk – for instance, via a zero-hours contract or other flexible work arrangement – will be conducive to lower production costs for her employer, who will consequently be relieved of the ‘holdover’ cost of retaining superfluous labour during relative product market downturns. Seeking to impose such an arrangement on the worker will therefore likely be in the employer's interest. However, in response the rational worker will (in theory) ask for corresponding contractual compensation, such as a higher rate of pay to compensate her for the reduced risk-adjusted value of her human capital, relative to what it would have been worth under an orthodox guaranteed-hours employment contract. At least, these are the distributive outcomes that we might expect to observe in the presence of idealised bargaining conditions: that is to say, full information and rational foresight on the part of the parties concerned, with no material disparities in their relative bargaining power.Footnote 73 However, once we move away from those (unrealistic) neo-classical assumptions, and introduce the real-world phenomena of bounded rationality, informational incompleteness, uncertainty and material bargaining inequality into the contractual fold,Footnote 74 a rather different picture emerges.Footnote 75
In the former of the two hypothetical scenarios set out above, the worker who trades off her ‘on the job’ freedom in return for a higher rate of pay will likely be unable to foresee accurately the full ambit of tasks that her employer might seek to impose on her in future. Indeed, inherent in the very legal nature of employment – as a relatively complex and diffuse relational contract – is the likelihood that many of the precise obligations of the parties (and especially the employee) will fall to be determined on a post-contractual basis under varying factual circumstances.Footnote 76 It is thus unrealistic to expect a worker to be able accurately to ‘price’ her expected level of commitment pre-contractually. In the majority of instances, though, it is arguably possible for the prospective worker to at least make a proximately accurate assessment of the likely nature and extent of her potential work tasks at the pre-contractual stage, such that ex ante pricing of her future human capital commitment – while by no means perfect – is nonetheless not necessarily frustrated as such.Footnote 77
By contrast, the latter of the above two scenarios presents considerably greater difficulty from an ex ante pricing perspective. The problem is that, under real-world bargaining conditions, pre-contractual pricing mechanisms tend to be systematically ineffective as a means of compensation for residual business risk exposure. This is principally on account of the inherent uncertainty involved in quantifying the degree of potential downside risk involved. To take the above zero-hours contract scenario as a pertinent example: in a volatile product (and, concomitantly, labour) market environment the on-demand worker's potential future return on human capital could vary considerably, and could even be non-existent for significant periods of time. Under such turbulent conditions, determining an adequate level of ex ante (eg wage-based) compensation for the notional worker is thus rendered an intractable task. There consequently exists considerable scope for unilateral opportunism, insofar as the employer enjoys leeway to provide an ex ante level of compensation that is insufficient to compensate the flexible worker for the full extent of residual business risk to which she is ultimately exposed. The result is an uncompensated reduction in the latter's risk-adjusted rate of return on human capital.
Insofar as the employer's gains from the zero-hours work arrangement (in the form of reduced holdover costs) are won merely at the direct expense of the worker in the above way; it follows that, from a normative (economic) standpoint, the arrangement between the parties will be conducive to no net added value for society.Footnote 78 However, even to the extent that adoption of a zero-hours work arrangement is conducive to a net social gain (in the sense that the employer's marginal benefit from the arrangement exceeds the worker's corresponding marginal cost), the ensuing outcome will still be socially sub-optimal if the real-world distribution of the surplus between the parties is significantly out of kilter with what it would have looked like in the absence of material uncertainty and relative power disparity.Footnote 79
Of further concern, though, is the longer-term social cost of such a lopsided risk distribution in reducing the level of continuing trust that the worker is willing to repose in the employer, as manifested by the former party's willingness to commit her labour to the latter's cause beyond the most basic level elicited by her ex ante contractual obligations.Footnote 80 This is because zero-hours and other divisible work arrangements lack the element of reciprocal ‘give and take’ that is implicit in the orthodox employment relation, in the sense that a fixed-wage employee is liable at various times to be both under- and over-compensated for their labour depending on prevailing market conditions, with the general expectation being that the parties will more or less break even over time. By contrast, on-demand (and especially zero-hours) work arrangements typically provide no credible assurance that such long-term ‘evening-out’ will occur. Consequently, the on-demand (and especially zero-hours) worker will likely be discouraged from rendering herself vulnerable to employer exploitation by unilaterally assuming any (extra-minimal) level of work commitment that is liable to go uncompensated.Footnote 81 The outcome is a downward spiral in employer-worker trust; resulting in reduced worker productivity, higher employer production costs and, ultimately, a mutual net loss for the parties concerned.
4. The structural similarity between flexible work and unsecured financial investments
Furthermore, it can be said that the position of the flexible worker, as an at-risk human capital investor, is in many ways structurally similar to that of an at-risk or unsecured financial investor. This is because, in an orthodox (ie guaranteed-hours) employment arrangement, the employee's continuing contractual assurance of minimal working hours is the functional analogue of collateral security in the case of debt finance. That is, insofar as it provides an ex ante buffer for the employee's committed human capital against exogenous market shocks, at least for the duration of the relevant employment relation.Footnote 82
Indeed, the limitations of pre-contractual (eg price- or wage- based) measures as effective compensation for vulnerable or ‘at-risk’ investments have been well-documented within law and finance literature.Footnote 83 In particular, it has been explained how fixed-interest debt financing is an inherently unsuitable means of funding firm-specific or non-redeployable assets, whose insusceptibility to liquidation renders them ineffective for the purpose of collateral security.Footnote 84 It follows that the creditor in this situation is exposed more or less entirely to the residual risk of debtor firm failure, which is impossible to quantify and price reliably by way of an ex ante contractual interest rate.Footnote 85 Accordingly, the lender may instead have recourse to ex post ‘settling-up’ mechanisms such as ‘event-of-default’ clauses, which will function as a surrogate for comprehensive pre-contractual risk-pricing by enabling the creditor to revise certain aspects of her initial arrangement with the debtor (or potentially even terminate it outright) in response to specified post-contractual variations in the former party's initial risk exposure.Footnote 86
Alternatively, in underwriting at-risk (ie contractually unsecurable) firm investments it may even be prudent for the relevant funder to dispense with debt finance altogether and instead acquire effective part-ownership of the firm itself by means of equity finance. From this perspective, equity capital could be regarded as the ultimate form of ex post settling-up mechanism insofar as it enables an at-risk investor to dispense entirely with the right to an ex ante fixed contractual rate of return on her capital. In its place is substituted: (i) the entitlement to appropriate a share of the firm's ‘upside’ residual profit as the effective price for assuming exposure to a portion of its ‘downside’ residual risk; and (ii) the supplementary entitlement to exercise a degree of control over the firm's management commensurate with the proportion of residual risk assumed. In this way, the at-risk financial investor is able to mitigate her susceptibility to post-contractual exploitation by bargaining for quasi-ownership rights over the investee firm itself, which theoretically enables her to influence directly the risk-adjusted rate of return on her committed capital on a continuing basis.Footnote 87
It therefore follows that, just as the absence of effective security (eg in the case of a firm-specific, non-redeployable investment) is liable to elicit the need for compensatory post-contractual protections (including quasi-ownership rights) on the part of an at-risk financial investor, so too is the absence of contractually-assured working hours prone to engender the same in the case of a flexible worker (as an at-risk human capital investor).Footnote 88 In particular, since – as explained above – ex ante pricing mechanisms (eg fixed contractual wage rates) are technically incapable of compensating at-risk human capital investors for their residual business risk exposure and ancillary costs,Footnote 89 there arises the corresponding need for surrogate ex post settling-up mechanisms in the case of on-demand and other flexible workers. Such measures must be designed to facilitate effective post-contractual responses by workers to attempted adverse variations in their risk-adjusted rate of return on human capital – the aim being to bring real-world distributional outcomes of flexible work closer to the notionally optimal outcomes that would ensue under idealised bargaining conditions, where accurate pre-contractual pricing of risk exposure is technically feasible.
Indeed, insofar as flexible workers are – like underwriters of unsecurable financial investments – exposed more or less entirely to the (contractually-irreducible) residual business risk of the enterprise in which their capital is invested, it follows by analogy that flexible workers will only adequately be compensated for the full extent of their commitment to the employer firm by means of outright quasi-ownership rights therein. Ideally, such rights should include: (i) the entitlement to compensatory ‘upside’ profit-sharing rights commensurate with their downside residual risk exposure, by means of a cooperative firm ownership structure (‘co-op’)Footnote 90 or employee share ownership plan (ESOP);Footnote 91 and also (ii) the supplementary entitlement to influence the firm's overall business strategy and risk appetite by means of employee information and consultation (ICE) rights,Footnote 92 partial board representation (or ‘co-determination’)Footnote 93 and/or other high-level managerial decision-making rights.Footnote 94
Accordingly, via profit-sharing rights flexible workers can be effectively compensated for their residual risk bearing function vis-à-vis the firm. Furthermore, via co-decision-making rights in the firm, flexible workers will formally be in a position to exert influence over some of the key variables determining their prevailing rate of return on human capital.Footnote 95 Under the idealised conditions of the above hypothetical bargaining model it is thus possible to envisage a dual paradigm of work emerging, whereby human capital suppliers opt either for the relative contractual insulation from extraneous risk that an orthodox employment arrangement provides, or, alternatively, for the lesser insulation but correspondingly greater quasi-proprietary entitlement that a flexible work arrangement offers. In this way, prospective workers would be in a position rationally to weigh up the respective costs and benefits of each type of arrangement in determining which particular trade-off is more amenable to them. Meanwhile, employers would be significantly constrained as regards their capacity to offer on-demand and other flexible work. This is because such arrangements would only rationally be assumed by workers where employers are willing to cede to workers the level of proprietary and governance influence that is needed to make flexible work mutually beneficial, at least in comparison to orthodox guaranteed-hours work arrangements.
Of course, it is one thing to claim that such a dual structure of work is notionally ideal or optimal from an economic efficiency perspective, and quite another to suggest that it is likely to emerge in practice, at least without some form(s) of mandatory regulatory compulsion. Indeed, it is strongly arguable that the likely persistence of the various real-world obstacles to efficient employer-worker bargaining alluded to above necessitates the advancement of a compelling normative case for the effective ‘correction’ of what is fundamentally a structural labour market failure: that is, the inherent incapacity of most flexible or non-standard workers to negotiate for effective proprietary and/or governance compensation for their residual risk bearing function within the firm.Footnote 96 The task of identifying the specific regulatory measures that might be conducive to fulfilling this corrective function is beyond the scope of the present paper. For immediate purposes, though, the recent UK Labour Party proposals to require: (i) that companies employing at least 250 persons progressively allocate up to 10% of their share capital to a special ‘inclusive ownership fund’ (or IOF) for the collective benefit of their workforce;Footnote 97 (ii) that at least one-third of seats on the boards of directors of companies employing 250 or more persons are allocated to worker representatives;Footnote 98 and (iii) that worker pre-emption rights are created vis-à-vis shares in companies that are subject to prospective acquisition,Footnote 99 would together seem like a constructive starting point for this discussion at the very least.
Conclusion
Irrespective of one's personal opinion on the social desirability (or otherwise) of the gig economy and flexible work practices in general, at least one thing would appear certain: both for legal scholars and social commentators more broadly, this topic is unlikely to recede into the shadows any time soon. At the time of writing, the widely-reported Uber v Aslam caseFootnote 100 on the proper legal classification of gig economy operatives awaits the next stage of its hearing before the UK Supreme Court,Footnote 101 as similar litigation involving other well-known gig platform operators likewise rumbles on.Footnote 102 In the policymaking realm, meanwhile, the UK government looks set to implement many of the recommendations of the 2017 Taylor ReviewFootnote 103 and House of Commons Committees’ ReportFootnote 104 on modern employment practices,Footnote 105 which will probably have significant repercussions for the future shape and scope of the UK's gig-based and on-demand labour market sectors.Footnote 106 And, even in the traditionally staid and prosaic world of UK corporate governance regulation, landmark change has recently occurred in the guise of the new (as of July 2018) UK Corporate Governance Code recommendation for the involvement of workers – and, in particular, casual and other non-standard members of the workforce – in the governance of premium-listed companies.Footnote 107
Since the ongoing judicial and regulatory developments in this field are likely to be subject to the intense scrutiny of both employment law and corporate governance specialists for the foreseeable future, the present author has opted to refrain from venturing into this particular field of enquiry. Instead, this paper has been designed for the purpose of bringing alternative inter-disciplinary insights, which have traditionally been more at home within the field of corporate law, to bear on the notionally separate factual domain of contemporary employment practices. Indeed, although corporate and employment law are – academically speaking at least – two traditionally autonomous fields of enquiry with relatively limited crossover of ideas and insights,Footnote 108 this belies the reality of how the industrial relations world operates. In practice, the significant and complex interactions between a firm's business-organisational structure and work practices mean that, for scholars and students on either side of the artificial corporate/employment law divide, there is much to be gained from the mutual sharing of ideas and perspectives.Footnote 109
Above all, it is the present author's view that many of the core arguments and insights from law and economics which are familiar to corporate law scholars are applicable just as readily to the employment law and industrial relations context. Moreover, this stands true nowhere more so than in the field of flexible or non-standard work practices, where some key insights from the economic analysis of law – and, in particular, the Coasean theory of the firm – have a great deal to add to the existing body of received legal knowledge and opinion.
This paper aspires to provide additional conceptual fuel for continuing scholarly and policy debates in the field. In particular, it is hoped that the above insights will help to equip those intuitively uncomfortable with the social outcomes of many controversial modern work practices with a broader array of normative arguments on which to draw in expressing their opposition and key concerns. At the same time, as is the case when critiquing any innovation or phenomenon that holds out the potential for significant socio-economic advancements, one should remain alert to the risk of throwing out the baby with the bathwater.Footnote 110
