Learning in Standard Form Contracts: Asymmetric vs. Symmetric Terms, Study notes of Literature

Download Learning in Standard Form Contracts: Asymmetric vs. Symmetric Terms and more Study notes Literature in PDF only on Docsity! LEARNING IN STANDARD FORM CONTRACTS: THEORY AND EVIDENCE Giuseppe Dari-Mattiacci and Florencia Marotta-Wurgler* November 27, 2018 ABSTRACT Why are some contractual terms revised continuously while others are stubbornly fixed? We offer an account of both change and stickiness in standard form contracts. We hypothesize that drafters (sellers) are more likely to revise the terms they offer when they have an opportunity to learn about their value from experience. Consider a warranty. Offering a warranty in an initial period will expose sellers to claims about malfunction by purchasers, allowing sellers to learn whether it is desirable to offer it going forward. When drafters are unable to learn in this manner, either because they fail to offer such learning-enabling terms initially, or because the term in question is one where there is no increased opportunity to learn, such terms will be revised less frequently. While learning and change occur through various channels, we posit that terms that carry an opportunity to learn from experience will be revised more frequently, where terms or term modalities that do not will contribute to stickiness and stagnation. Our results support this hypothesis. Using a large sample of changes in consumer standard form contracts over a period of seven years, we find that sellers are more likely to revise terms that offer an opportunity to learn from experience than those that do not. The results suggest that standard form contract terms evolve over time as sellers learn experientially about their costs and risks. Our results have normative implications for the optimal design of default rules. JEL classification: K12. Keywords: standard form contract, boilerplate, evolution of contracts, learning. * Giuseppe Dari-Mattiacci: University of Amsterdam School of Law and Amsterdam Business School. Florencia Marotta-Wurgler: New York University School of Law. The authors would like to thank Lisa Bernstein, Steven Choi, Kevin Davis, Marco Fabbri, Clayton Gillette, Ron Gilson, Scott Hemphill, Gerard Hertig, Keith Hylton, Henrik Lando, Mark Ramseyer, Alan Schwartz, Robert Scott, David Webber, Kathryn Zeiler, Jonathan Zitnyck, and the participants in the Conference on Contractual Black Holes (April 7th and 8th, 2017) at Duke Law School, the annual meeting of the European Association of Law and Economics (September 14th – 16th, 2017) in London, and seminars at NYU Law School, ETH Zurich, the Institute for Information Law at the University of Amsterdam, the faculty workshop at Boston University School of Law, and the 2018 American Law and Economics Association annual meeting, for helpful comments and suggestions. DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS 2 1.! INTRODUCTION One of the defining characteristics of standard form contracts is a high degree of standardization. Another is that their terms tend to be “sticky.” In theory, contracting parties should revise their agreements when doing so enhances the value of their transaction. However, the literature has identified a number of factors that might reduce contracting parties’ incentives to deviate from the norm or default rules, even when alternative arrangements enhance the value of the transaction. Yet, stickiness is not a general phenomenon. Some terms seem to be resistant to change, like pari passu clauses in sovereign bond agreements,2 while others get updated very quickly. We propose and examine empirically a novel account of both stickiness and change in standard form contracts. We begin from the reasonable premise that contract drafters may be initially uncertain about the net value of a particular term. Over time, however, drafters are able to learn the relative costs associated with such terms, leading them to drop some while adding or revising others. Firms learn in many ways, including learning about the terms offered by competitors, cases litigated in court, technological innovations, and news reports, among others. A common feature of these learning channels is that they tend to function largely independently of the specific contractual choices firms make.3 Contrast this with experiential learning, where firms learn directly from experience with and feedback from consumers and contracting parties. That is, there are certain terms whose relative value is best ascertained through the direct feedback only their use generates. When learning is experiential, the firm’s ability to learn depends on its past contractual choices.4 This is the focus of our paper. Consider a default implied warranty. A firm may contemplate including a disclaimer of implied warranties in its standard form contract. Offering the warranty may allow the firm to charge a higher price for the product but it will also expose the firm to some costs due to consumers claiming a remedy. During the time the firm must make its choice, the extent to which the costs outweigh the benefits of the warranty to consumers may be uncertain given that these values depend on factors that are harder to ascertain ex ante, such as on the frequency of product breakdown, the types of consumer losses, and the frequency at which consumers claim a remedy. If the firm offers the default implied warranty, it will risk exposure to future financial liability, but would also acquire a valuable opportunity to learn the true costs of the warranty and inform future choices. Opting for the disclaimer saves costs in the short run but also prevents the firm from learning. That is, the firm’s choice as to the mode of this particular term (opting into the warranty default versus opting out) affects its ability to experience and learn the term’s net value. 2 Gulati & Scott, id (explaining contracting parties’ reluctance to revise pari passu clauses in sovereign bond agreements after unfavorable interpretations by courts). 3 For a review of the literature on learning and innovation in the standard form contract setting, see Section 2. 4 Our approach is related to the vast literature on strategic experimentation (Bolton and Harris). DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS 5 chose to include the term in its learning mode at an initial period. We also find that asymmetric learning terms adopted in their learning mode are more likely to be revised than symmetric learning terms (and the reverse for asymmetric learning terms adopted in their non-learning mode). The results are statistically significant and robust. The paper proceeds as follows. In Section 2, we review the literature on standard form contracts and contractual innovation. In Section 3, we propose a theory of experiential learning and derive testable predictions. In Section 4 we test these predictions using a unique dataset of standard form contracts. In Section 5, we conclude with additional considerations on the normative implications of our theory. The Appendix contains all the details of the empirical analysis and the proof related to the theoretical model. 2.! LEARNING, STICKINESS, AND INNOVATION IN STANDARD FORM CONTRACTS 2.1.! Stickiness and change The benefits of standardization are well understood and have been explored extensively in the literature. As terms become increasingly common and well-known, they are easier for contracting parties and courts to interpret. They also confer various spillover effects, such as lower reading costs, increased certainty of legal interpretation, and reduced litigation risk (Kahan and Klausner; Choi and Gulati; Gillette). Yet these, and other, benefits resulting from the use of boilerplate may stand in the way of change, even when doing so might be efficient (Kahan and Klausner). Reluctance to change in light of a superior alternative could give rise to agreements with terms that no longer serve the contracting goals of the parties, either because they no longer reflect the optimal allocation of rights and risks between them, or because they might be interpreted unfavorably by a court, among others. A number of factors contribute to stickiness. Stickiness tends to be associated with markets that experience network benefits that arise from firms’ simultaneous adoption of a term (Kahan and Klausner), as well with agreements drafted by law firms, whose hierarchical structure and tendency to re-use old forms (Gulati and Scott; Hill). Firms’ incentives to innovate are further diluted by weak property rights in contractual innovations (Davis) and the existence of default rules. When states enact particular defaults, contracting parties might find it cost effective to just adopt them (Goetz and Scott; Schwartz and Scott (1995; 2003)), a tendency further reinforced by the status quo bias (Korobkin). Such parties might also be reluctant to deviate from them when they perceive that opting out might signal negative information, even if value generating (Spier; Johnston; Ben-Shahar and Pottow; Bernstein; Kahneman, Knetsch, and Thaler). Despite these obstacles, change and innovation can and does still happen. Large repeat players, such as law firms and investment banks, can find it profitable to invest in innovation— even in the absence of strong property rights—through their ability to spread costs among clients DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS 6 (Kahan and Klausner; Gulati and Scott). In-house counsel in legal departments of firms engaged in mass-market commerce work closely with management and understand changes in technology that might give rise to new terms. In-house counsel are also more likely to receive feedback from offering or refraining to offer particular types of terms, allowing them to revise the agreements to adapt to new legal and market environments (Macaulay; Triantis). Finally, change and innovation can be spurred by exogenous “shocks,” such as new laws, changes in legal interpretations of terms, or technological advances, driving firms to revise their agreements. Most of the empirical evidence on contract change, innovation, and stickiness comes from studies of bond covenants and financial products. Kahan and Klausner, and Choi and Gulati (2006), found evidence of switching and learning costs in the corporate bond covenant context. Choi, Gulati, and Posner (2013) found an S-shaped innovation pattern in sovereign debt contracts, where parties slowly move from the old standard to a new one in response to various exogenous shocks. In the law firm context, Gulati and Scott found that lawyers in law firms failed to revise terms even after those terms had acquired ambiguous meanings that increased litigation risk. In the insurance context, Schwarcz (2011) found evidence of innovation away from the ISO form, which is the standard insurance document. Coates (2018) found significant changes in merger agreements over time, unveiling that such contracts had doubled in size, and that about 20% of such change were attributed to new terms. Finally, Marotta-Wurgler and Taylor (2013) found evidence of terms changing in reaction to litigated cases and changes in the enforceability of terms. To summarize, there have been numerous accounts to explain and document either stickiness or change in standard form contracts. In this paper, we propose a new mechanism to account for contract change: learning from experience. To the best of our knowledge, this is the first paper to explore this mechanism in the standard form contract setting and to offer an account explaining both stickiness and innovation in the absence of external shocks. 2.2.! Experiential learning versus other forms of learning Learning is a fundamental driver of change. Yet learning occurs in different ways. Firms can learn directly, by interacting with consumers, or through the experience earned from offering particular terms, or term modalities. Firms can also learn indirectly, in ways that are unrelated to the contractual choices made at an initial period. Consider a term in a EULA limiting the number of devices where the software can be installed, which can clearly affect demand for the product. Learning in this context occurs largely independently from the form of the contract term offered related to the number of devices. Rather, learning about consumer preferences and uses of software can be achieved by looking at purchasing patterns or examining the offerings of competitors in their own market and adapt terms accordingly. This type of learning takes multiple forms. Firms can learn from litigated cases about possible contractual choices as well as about the enforceability of particular clauses (e.g., a “change of terms” clause that allows firms to modify standard agreements unilaterally) (Gulati and DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS 7 Scott (2004); Marotta-Wurgler and Taylor (2013); Schwarcz). The literature on contractual innovation has also pointed out that firms can learn from each other’s contracts, creating a well- known free rider problem (Goetz and Scott; Davis). Law firms are also a conduit of indirect learning by transmitting knowledge to their seller-clients, who can then revise their terms accordingly. More recently, legal service firms like Bloomberg Law and Legal Zoom have begun offering standard terms for different types of contracts, allowing firms to innovate at a relatively low cost (Triantis 2013). Blogs, trade publications, word of mouth, and internet forums offer additional sources of free advice regarding terms. All of the aforementioned channels enable learning and change, but the mechanism by which this happens is unrelated to the firms’ experience with its adoption of a term or term mode. We refer to these forms of learning as “indirect” learning mechanisms because they can occur independently from the contractual choices made or direct interactions with customers. In other circumstances, learning is not indirect, but rather the result of firms’ interactions with consumers and, to some degree, the result of the products and terms offered. This form of learning is experiential in nature but the experience is not conditioned on the occurrence of an event that might implicate the contract. For example, firms can acquire valuable knowledge from interacting with consumers through employees and customer service channels, where learning isn’t necessarily mediated by the contract terms themselves (Hoffman).7 Consumers can call firms and inquire about the meaning or implications of a particular term, without demanding rights under the contract. This might allow firms to learn about how consumers understand the contract or particular features of a product or service. Firms also gain valuable information from feedback offered through online customer reviews (Ghose and Ipeirotis), which can lead to change. We refer to these forms of learning as “direct” learning, because they result from customer interactions or aspects related to the product or term offered. One particular form of direct learning is experiential with respect to the contract. In particular, firms can learn from experience resulting from the use and implication of a particular term. The most natural example is a warranty. A seller can offer it or not; if the seller offers the warranty and the product breaks, the consumer can bring a claim for breach of warranty. In honoring the warranty, the seller learns the costs of offering this particular term. Learning occurs if three conditions are met: the seller offers the warranty, the product breaks down, and the buyer brings a claim for breach of warranty. Note that the difference with this form of experiential learning as compared to other forms of direct learning, such as when consumers provide feedback, is that additional learning only occurs when a particular term or term mode is offered (and implicated) and not otherwise. There are some terms that tend to lend themselves to experiential learning more than others. Warranties, terms offering maintenance and support, and terms where firms get to experience the cost of offering them from consumer claims or actions tend to fit well 7 See David Hoffman, Relational Contracts of Adhesion, 85 U. Chi. L. Rev. (forthcoming 2018). DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS 10 consumer demand or for competitive reasons. Still, a firm that offers a warranty at an initial period receives additional learning as a result of its experience with consumers. As a result, such firm would be marginally more likely to revise it at a later period as a result of such learning. The warranty is an asymmetric learning term; more precisely, this is a term where the firm learns only if it offered the default option. Table 1.! Information-types and modalities of contract terms Information-type Default Opt-out Symmetric-learning terms (S) Learning Learning Asymmetric-learning terms (A) From default Learning Nonlearning From opt out Nonlearning Learning Table 1 provides an overview. Terms of type S are symmetric-learning terms and are such that the firm receives new information at time 1 irrespective of whether it adopted the default or opted out of it at time 0. An example of such terms are restrictions-on-use terms, described earlier. The distinctive characteristic of a symmetric term is that a decision of whether to revise a term at a later period will arise irrespective of a term’s mode at an initial period. A symmetric learning term might be revised or not, but that will occur independently from the experiential learning mechanism that we focus on. In contrast, terms of type A are asymmetric-learning terms: between time 0 and time 1, the firm learns the costs associated with the term only if it has adopted what we call the learning mode of the term, which could be either the default (as in the implied warranty example above) or the opt out. If the firm does not offer the learning mode at time 0, it will not learn anything—or, more generally, it will learn less—from interacting with consumers. To anticipate the central take-away from the model introduced below, consider the following simplified example. Assume that consumers value the default term at v and that the default could cost the firm either cH > v with probability p or cL < v with probability 1 – p. For simplicity, assume that ! = #$%#& ' and that the costs and value of the opt out are zero. With symmetric learning, the firm’s decision at time 0 is purely driven by a balance of expected costs and hence will choose the default if v > pcH + (1 – p)cL, that is, it will choose the default ( < (* ≡ , ' (low expected costs) and opt out otherwise (high expected costs). At time 1, adopters may discover that costs are high—this happens with probability p—and decide to switch to the opt out. The area of the grey triangle in the upper-left corner of the graph in Figure 1(a) depicts the ex ante probability mass of switches from adoption of the default term to opt-out. Conversely, firms with p > p opt out at time 0 and decide to switch with probability 1 – p. The grey triangle in this part of the graph depicts the ex ante probability of switches from opt out to default. Consider now asymmetric learning from the default option. If the firm chooses the default DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS 11 at time 0 it will face expected costs equal to pcH + (1 – p)cL but will also have an opportunity to learn and be able to switch. After learning, the firm’s time 1 payoff is v – cL with probability 1 – p (the firm learns that costs are low and confirms the time 0 choice) and zero with probability p (the firm learns that costs are high and switches to the opt out). This is a real-option value of the default option and pushes the firm to choose the default for its dynamic gains even though it yields static losses. Now the firm chooses the default if v + (1 – p) (v – cL) > pcH + (1 – p)cL, which yields a higher threshold for opting out (- ≡ ' .. We can visualize the firm’s choices in Figure 1(b). Compared to the symmetric learning term, more firms choose the default at time 0. Adopters, however, switch to opt-out with relatively high probability, especially in the range [pS, pA], that is, in those cases that would have resulted in opt-out at time 0 had the term been of a symmetric type. These are instances in which the default has a negative expected value and it is chosen purely for its learning value, that it, for the option to make a perfectly informed decision at time 1. Instead, Firms that choose to opt out irreversibly confirm that decision at time 1 because no new information is acquired in the meantime. Figure 1.!Adoption decisions at time 0 and time 1 (a) (b) 3.2.! Setup of the model In the model there are two sets of players: (i) a monopolistic firm of type p∈[0, 1], randomly drawn from the cumulative distribution function F(p), and (ii) a population of homogeneous consumers. At time 0, the firm draws its type and offers all of its consumers the same standard form contract, which can come in two guises. The firm can either adopt a default term prescribed by the law or opt out of it. In the interim period, the firm has an opportunity to learn the costs associated with DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS 12 the terms it offered. Then, at time 1, the firm can offer the same contract as at time 0 or revise it. The default term and the opt out have different values for consumers: let v be the value of the default relative to the opt out. Similarly, they entail different costs for the firm: let c∈{cL, cH} be the cost of offering the default term relative to the opt out, with c = cH with probability p and c = cL with the complementary probability 1 – p. (Note that this is equivalent to normalizing the value and cost of the opt out to zero and is without loss of generality.)10 Accordingly, the firm’s type p is such that, when offering the default, high types face high expected costs and low types face low expected costs. To capture the interesting scenarios, we focus on cases where cL < v < cH. The firm maximizes v – c; with full information, the firm should offer the default term if c = cL and opt out if c = cH.11 However, while v, cL, cH, and p are known to the firm, the cost c is unknown at time 0 and may or may not be known at time 1. Hence, time-0 choices are made in conditions of imperfect information. Depending on the choices made at time 0, the firm may experience the costs associated with different clauses and learn. We postulate that the firm learns c with probability λD ∈ [0,1] if it has adopted the default at time 0 and with probability λO ∈ [0,1] if it has opted out. The pair (λD, λO) is term specific and can interpreted as the information-type of a term. Terms characterized by λD = λO expose the firm to symmetric learning. In contrast, terms characterized by λD > λO yield asymmetric learning, with the default offering more learning opportunities. That is, the default characterizes the learning mode of the term. Similarly, if λD < λO, the opt out characterizes the learning mode.12 Note that the model allows for various degrees learning and of asymmetry in learning. Before examining the firm’s contractual choices, let us enrich the model with three additional ingredients. First, firms that opt out of the default term may incur an opt-out cost k ≥ 0 10 The normalization is without loss of generality More precisely, v = vD – vO, where vD is the consumers’ willingness to pay for the product if the contract includes the default term and vO is their willingness to pay in case of opt out. Similarly, cH = cH D – cH O is the differential in costs when the term involves high costs and cL = cL D – cL O when costs are low. Note that we do not make any assumption as to the relationship between cL D and cH D and similarly for cL O and cH O: costs are high or low only in a relative sense. We only assume that cH > cL, implying that terms of type H should be offered in their default version while terms of type L should be offered in the opt out version. Yet, this is an innocuous labeling choice and assuming the opposite would only require us to change the interpretation of the results. Note also that our setup allows for situations where cL D = cH D while cL O ≠ cH O. In this case a firm offering the default option experiences costs without being able to infer whether the term is of type H or of type L: hence the firm offering the default does not learn c. In contrast, a firm offering the opt out would learn because cL O and cH O are different. This is a case of asymmetric learning from the opt out option and illustrates our framework where firms experience costs but may or may not learn. The same is true for learning from the default option only. 11 Since a monopolist can charge homogeneous consumers for their willingness to pay, the firm can extract v through the price. If v ≥ cH the firm offers the default irrespective of whether costs are high or low. Similarly, if v ≤ cL the offers the opt out irrespective of costs. These are uninteresting cases for our purposes because (the lack of) information plays no role in the firm’s choices and hence there is no scope for learning. 12 See note 10 for a formal implementation of the asymmetry in learning. DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS 15 Δ (∗ ≡ λ9(CD ( E∗ 8 Similarly, the mass of switches away from the opt out to the default at time 1 is equal to Ω (∗ ≡ λ; 1 − ( CD ( , E∗ We will now analyze how the firm’s decisions depend on the information-type of a term, @9, @; and demonstrate two general results (all proofs are in the Appendix). Proposition 1.! The learning mode of a term is chosen increasingly often at time 0 as learning becomes more asymmetric. Absent learning, expression (4) would be a ratio of the net value of the default (v – cL) over the sum of the net values of the two options (v – cL + cH – v = cH – cL), weighed by the firm’s time-0 and time-1 sales (1 + w). With learning, additional terms enter the expression, which capture the real-option value of choosing either version of the term. The marginal type p* increases with λD, enlarging the set of firm types p < p* that choose the default at time 0. Similarly, p* decreases with λO, expanding the use of the opt out. As learning becomes more asymmetric—that is as'λD – λO(increases—the learning mode yields learning relatively more frequently and hence becomes more attractive for firms at time 0. Proposition 2.! Switches away from the learning mode of a term occur increasingly often at time 1 as learning becomes more asymmetric. If learning is asymmetric, we know from Proposition 1 that the learning mode of the term is chosen more often at time 0. This effect expands the set of firms that can potentially switch to the other mode at time 1. In addition, those firms learn relatively more often—precisely because they have chosen the learning mode—and hence more of them will switch at time 1. 3.4.! Other determinants of contractual choice We now explore how a firm’s contractual choice is affected by the costs of opting out of the default option at time 0 or at time 1, the firm’s time-1 growth prospects and switching costs at time 1. Opt-out costs. Opt-out costs k add to the value v of choosing the default and make it less likely that a firm will opt out of it both at time 0 and at time 1. However, if the firm learns in the interim period, opt out costs are irrelevant to the firm’s choice as long as cL < v + k < cH (Assumption 1). If the firm does not learn, the firm simply confirms its time-0 choice irrespective of k. Therefore, opt-out costs are felt particularly at time 0, when the firm chooses under imperfect information, and make the choice of the default option more likely. In turn, this expands the set of firms that, DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS 16 having chosen the default, may learn and switch to the opt out option at time 1. Proposition 3 formalizes these intuitions. Proposition 3.! If opt-out costs increase, then the default term is chosen more often at time 0 and there are more switches away from the default at time 1. Growth prospects. Growth prospects have two effects: they magnify the (dynamic) gains from learning but also magnify the (static) net value of a term. Hence, an increase in growth prospects may or may not result in more learning depending on which effect dominates. To elaborate, the two modalities of a term may or may not be symmetric with respect to their net values. If ! = #$%#& ' , we have v – cL = cH – v (the net values of the default and the opt-out are the same) and the firm’s contractual choices are driven purely by learning. In this case, growth prospects stimulate learning: firms with greater growth prospects choose the learning mode of a term more often at time 0 and hence also switch away more frequently from the learning mode at time 1. However, if the former condition does not hold true, we may have v – cL > cH – v (or vice versa). In this case, growth stimulates learning only if the net value of the learning mode is large enough. Otherwise, the result is reversed. Proposition 4 provides a general formulation of these results. Proposition 4.! There is a neighborhood of ! = #$%#& ' such that, if the firm’s growth prospects increase, then the learning mode is chosen more often at time 0 and there are more switches away from the learning mode at time 1. This result is reversed if the net value of the learning mode is sufficiently small. An obvious corollary to Proposition 4 is that in case of symmetric learning terms, both modalities of the term imply learning and hence the choice is purely driven by the term’s net values. In particular, greater growth opportunities result in a more frequent choice of the default option if v – cL > cH – v, and of the opt out if v – cL < cH – v. Switching costs. Switching costs add an implicit tax on learning, thereby making learning a less attractive option, restricting the choice of the learning mode at time 0 and, consequently, the frequency of switches away from the learning mode at time 1. This result, however, may be reversed if the learning mode also guarantees the smaller static net value, which, as above, weighs against learning. Proposition 5 formalizes these observations. Proposition 5.! There is a neighborhood of ! = #$%#& ' such that, if switching costs increase, then the learning mode is chosen less often at time 0 and there are fewer switches away from the learning mode at time 1. This result is reversed if the net value of the learning mode is sufficiently small. DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS 17 As above, Proposition 5 implies that in case of symmetric learning terms contractual choice is purely driven by the term’s net values. In particular, greater switching costs result in a more frequent choice of the default option if v – cL < cH – v, and of the opt out if v – cL > cH – v. 3.5.! Discussion of the main assumptions 3.5.1.! Market structure (other than monopoly) In the model we focus on monopolistic firms. Considering firms with less than full market power would not qualitatively alter the results. Some degree of competition would reduce the firm’s ability to capture consumer surplus thereby requiring us to distinguish between the (relative) value consumers attach to the default, v, and the price increase that the firm is able to sustain when offering the default, which could be less than v when firms compete. This, however, would not alter our analysis, as we allow v to vary. However, in a fully competitive market, prices track costs, not consumer surplus. Hence firms might adjust the price they charge to consumers after learning the costs of different clauses. These adjustments may erode firms’ profits but should not affect the key mechanism behind our model: firms would still be induced to offer the cheapest option, which is unknown at the outset in our model. Yet, a formal model of standard form contracts in competitive markets might unveil additional implications. Competing firms might also learn from each other, which both boosts learning—because it magnifies the effects of any individual firm’s experimentation with new clauses—and hinders it—because it creates a free-riding problem that reduces a firm’s incentives to experiment. While this aspect of the problem would add a layer of complexity to the analysis, it would not affect our basic distinction among contract terms based on their learning characteristics and hence would not qualitatively alter our results. Finally, competitive forces might induce firms to follow what most of their competitors do because, for instance, consumers might be unwilling to buy a product that is offered together with an unfamiliar set of clauses. We already consider in the model the costs of opting out of the default term. A similar analysis could be applicable to the case of opting out of the industry standard terms. 3.5.2.! Heterogeneous consumers, tailoring and screening Our framework applies to cases in which firms offer standard form contracts to consumers. Yet, firms routinely attempt to tailor their contracts to the specific characteristics of individual or groups of consumers. Firms in our model offer standard form contracts to consumers, but costs associated with offering a specific term may vary with consumer characteristics. Moreover, different consumers may value the same term (say a warranty) differently. In these cases, firms might find it advantageous to tailor contracts to specific consumers or consumer groups rather than offering DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS 20 probability of learning (because of more limited feedback). While this is an interesting extension of our model it is unlikely to affect our results: after the firm has set (λD, λO) our model would carry on virtually unchanged. In addition, the extent to which firms can treat similar consumers differently is limited both by law and by reputational concerns. 3.5.6.! Reasons not to revise a term after learning The theoretical framework presented above focuses on learning direct costs, but the choice of terms can generate other forms of learning that affect contractual choices at a later stage. Consider, for example, a retailer that sells products manufactured by a number of suppliers and is uncertain about the quality of the products of each supplier. Offering a secondary warranty to consumers could be a way to obtain feedback on the quality of the firm’s suppliers. If the product breaks down frequently, the firm learns that its supplier delivers low-quality products. The interesting implication is that, in this case, the firm's response to learning is a change of supplier rather than a change of term. The firm may want to keep offering the warranty in order to learn about the new supplier. We do not elaborate on this alternative learning motive but we stress that this is also a form of experiential learning. Conversely, a choice of law clause may or may not be desirable depending on whether it lowers or raises the costs of litigating a case in court for the firm depending on unknown factors, determining whether the firm faces high or low costs. Learning about these costs may induce the firm to amend the clause at a later time. Our analysis applies to these cases. 4.! EMPIRICAL ANALYSIS We now put our theory to bear on the contractual choices made by real firms. We first derive empirically testable predictions from our theory. Next, we present our dataset and empirical results. 4.1.! Empirical implications of the theory While it is difficult to disentangle empirically the reasons behind firms’ adoption of terms in an initial period (given the multitude of factors likely affecting such decisions, many of which are hard to measure), examining firms’ decisions to revise such terms at a later period can offer some interesting insights regarding possible drivers of contractual choice. We explore learning from previous contractual choices against the attractiveness and stickiness of default terms. Limitations in data availability allow us to test only a subset of the model’s predictions following from Propositions 1 to 5, which we restate below. Prediction 1. The probability that a firm will amend an asymmetric-learning term at time 1 is DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS 21 higher if the firm has chosen the learning mode at time 0. The firm’s decision to revise an asymmetric-learning term is largely affected by the firm’s choice at time 0. Adopting a term in its learning mode at an initial period allows the firm to re-evaluate past contractual choices and amend them if new information suggests that a different choice is more advantageous. Prediction 1 also identifies a mechanism by which “black holes” could come about. If the firm has chosen a nonlearning mode at time 0, it will not see new information and might fail to revise the term in question at time 1. Inefficient or meaningless terms might survive due to the asymmetric nature of learning. In contrast, firms adopting the learning mode of the same term at time 0 stay away from them. Such “black holes” or pockets of inefficiency might affect only a portion of the firms in the market, especially when other learning channels play weak roles. Prediction 2. The probability that a firm will amend a symmetric-learning term at time 1 does not depend on the term chosen at time 0. Contrary to asymmetric-learning terms, here the firm’s initial choice does not affect the firm’s propensity to revise the term. For these terms, experiential learning, or learning from other channels, occurs (or not) irrespective of the contractual choice at time 0. If the firm learns from experience, learning will occur symmetrically from both the default and the opt-out option. We should observe revisions motivated by experience as well as other learning modes in this case but such revisions should be equally likely for firms that adopted the default and for firms that opted out of it at time 0. The same is true for when the firms learns from other means. Revisions of the term at a later date will be uncorrelated with the contractual choices made during the earlier period. Prediction 3. If default terms are inefficiently often chosen at time 0, default terms will be amended more frequently than non-default terms if they offer an opportunity to learn. Default contractual terms have long been recognized as important determinants of contractual choice. Implications of this observation come in two guises. On the one hand, if default terms are more frequently chosen, this could apply both at time 0 and at time 1. If, however, the choice of a term is largely determined by the term being a default, default choices at time 0 are more likely to result in inefficient outcomes. We anticipate that such defaults will be more likely to be amended at time 1 if the firm has had an opportunity to learn in the meantime. This effect should be visible both in symmetric and in asymmetric-learning terms. In the symmetric ones, the learning terms will be revised at time 1 more often towards the opt-out option if the default was inefficiently chosen at time 0. In asymmetric-learning terms, revision should be more frequent when the default is the learning mode than when it is the nonlearning mode. Both implications point to an important role of default contractual terms in determining firm choices going forward. If this is the case, switches at time 1 should be largely explained by the fact that a term is a default. This prediction will allow us to contrast defaults to learning as DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS 22 alternative explanations for change in standard form contracts. We turn to the empirical analysis in the next section. 4.2.! Data and methodology We test our hypotheses using a sample of software license agreements governing the use of pre- packaged software. End-User License Agreements (EULAs) typically present a rich set of standard terms; while the terms typically vary both across and within markets, EULAs follow a predictable structure (Marotta-Wurgler). This allows for meaningful comparisons across contracts. We examine the rate of change of terms from 2003 to 2010 in accordance with sellers’ opportunity to learn from their presence or absence. We use the sample of EULAs introduced by Marotta-Wurgler and Taylor (2013), which tracks the changes in the terms of EULAs found in typical “prepackaged” (i.e., non-customized) software products and compare their content in 2003 and 2010. That study examined the change in 32 EULA terms from 246 firms that sell their software on their corporate Internet sites, including large, well-known, software publishers, as well as smaller companies. For each of the companies, the dataset includes a representative product along with data on various market, product, company characteristics, and of course, the EULA both in 2003 and in 2010. For each EULA in each period, we tabulate the presence of 32 standard terms across seven categories of related terms, such as scope, warranties, limitations of damages, etc. We further classify each term into categories reflecting the extent to which offering a given term gives sellers an opportunity to learn directly from experience, either symmetrically or asymmetrically. We also take account of other factors that might affect firms’ decisions to revise terms at a later time, such as their size, age, and whether they have in-house counsel. 4.2.1.! Summary statistics Table 4 presents summary statistics. Panel A reports company characteristics for the sample firms. Average revenue in 2003 was $287.5 million and the median was $1.7 million. Average and median revenue in 2010 were $539.1 million and $2.2 million, respectively. The percentage of public companies grew from 11% in 2003 to 14% in 2010. The sample includes data on legal sophistication in 2010, proxied by firms’ choice of legal advice, including whether they have in-house counsel, at least one internal lawyer, or routinely hire outside counsel. All public companies are assumed to receive sophisticated legal advice. In total, 74% of firms for which these data were available received relatively intensive legal advice, which might affect firms’ propensity to revise terms at a later date. Panel B lists product and market characteristics in 2003 and 2010. The average price of the products in the sample was $812 in 2003 and $841 in 2010. Thirty-six percent of the products are oriented toward consumers or small home businesses, rather than large businesses. One percent of DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS 25 difference of 3.4 percent, while small, is statistically significant at the 10% level. 4.3.! Analysis We now explore the extent to which the changes reported in Table 3 are more likely depending on the initial choice of terms as well as when sellers have an opportunity to learn. The nature of our data prevents us from making any inferences regarding the initial choice of terms, as these are also likely the result of past contractual choices. We can, however, measure the extent to which default rules are predominantly chosen and measure the extent to which these are revised in a later period. Panel A in Table 7 begins by exploring the stickiness of default rules in the data by reporting the extent to which sellers chose to match the default rules of the UCC at the initial period as well as the probability of revising a term given their initial mode in the previous period. The top right figure shows that among 32 terms in total, and 8,448 EULA-term observations, 30.8% of all terms in 2003 were at the opt-out value, whereas the remainder, or 69.2%, matched the default rules, indicating a strong gravitational pull towards the default previously identified in the literature. Yet default terms are not set in stone. In 2010, the fraction of terms that match the default decreased to 66.7%. Indeed, 65.3% of all terms were at default values in both 2003 and 2010, but 3.9% were at default values in 2003 and opted out in 2010. In terms of probabilities, the right panel shows that the probability of changing a term in 2010 given that a term was in an opt-out and default value in 2003 was 0.045 and 0.056, respectively. The 0.011 difference is statistically significant at the 5% level. While terms are more likely to begin at the default, the probability that they will be revised at a later period is larger if the term starts at the default, offering support to the known view that sellers might be inefficiently choosing default terms in the initial period due to opt-out costs. With this baseline in mind, we test predictions 1 and 2 by dividing the data into whether the term generates symmetric or asymmetric experiential learning opportunities. Panel B presents data on symmetric learning. As noted earlier, sellers might be learning about these terms through other means, independent from experience and irrespective of whether the term matches the default rule or not. We have no a priori hypotheses as to how these additional sources may inform sellers. For our purposes, all we care is to know whether change is more likely to be associated with one mode of the term or the other. The results show that, again, defaults are powerful determinants of contract terms in the initial period. In this case 75% of symmetric terms match the default rule in 2003, only to change to 72.4% in 2010, indicating some change away from defaults. More interesting for our purposes, however, is the probability of change conditional on the starting point. Recall that we predicted that the starting point for these types of clauses would be a poor predictor of change. In fact, the probability of changing a term is precisely the same, or 5.2% depending on where the term is in 2003. DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS 26 Contrast this with Panel C, the results for asymmetric terms. In 2003, 64.2% of all such terms matched the default rules of the UCC, a number that shrank to 61.8% in 2010. The right panel shows that the probability of change for terms that matched the default in 2003 is 6.1%, in contrast to 4.2% for non-defaults. The difference is significant at the 5% level. Even for the asymmetric learning clauses—and consistent with the findings in Panel A examining all terms— terms are more likely to be revised when they start at the default rule, regardless of the learning mode. Once we divide asymmetric terms up into their learning modalities, a new picture emerges, as seen in the bottom panel of Panel C. In 2003, asymmetric terms are included in their learning and nonlearning modalities about equally. However, and in contrast to the symmetric terms, where the probability of changing a term was independent of the original allocation of the term between default and nondefault, in the asymmetric scenario, the original learning mode matters. The probability of changing a term given that the 2003 contract included such term in its learning mode is 0.072, in sharp contrast to the 0.034 that occurs when the term is not in its learning mode. They also support the prediction that asymmetric learning terms adopted in their learning mode at an initial stage are more likely to be revised than symmetric learning terms (7.2% vs. 5.2%, respectively); while the reverse is true for asymmetric learning terms adopted in their non-learning mode at the initial stage (with a 3.2% revision probability). The findings support the basic prediction that opportunity to learn helps explain contractual change and innovation. That being said, while the findings are consistent with our theoretical predictions, observational data can never prove that learning is one of the causes of changes in terms. Future work could complement the large-sample evidence with case studies and interviews with general counsel to give a nuanced, descriptive view of why standard terms change over time. Figure 2.!Probability of Term Change DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS 27 These findings are illustrated in Figure 2. The left bars show the probability of change conditional on their 2003 starting point (default versus opt-out). The bars are the same height, consistent with the mode of the term conferring no consistent learning advantage. Contrast this to the bars on the right. Change is more likely to happen if the terms are switched on their experiential learning modes in 2003, as opposed to their nonlearning mode. Table 5 reports ordinary least squares regressions including company, product, and market control variables. The first column just repeats the results from the bottom of Panel C of Table 4. The second column adds firm (contract) fixed effects, controlling for the overall propensity of a given contract to change. The fact that the coefficient on learning does not budge indicates that there is not a tendency for some firms to make wholesale changes to their policies, including their learning terms; a given learning term is equally likely to change “within” a contract whether the same firm is changing many or few other terms. The third and fourth columns show that the probability of changing away from a term at the default in 2003 is robust to the overall propensity to change the contract, but the effect is only half that of the probability of changing the term as a function of the term’s learning status, and is a distinct effect.21 Logit regressions yield very similar 21 The terms more likely to change when set in their learning modes are those that address the buyer’s ability to reverse engineer the product, and those that determine the sellers’ implied warranty obligations and liability for consequential damages of the buyer (unreported). 0.052 0.052 0.072 0.034 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 Symmetric Nondefault Symmetric Default Asymmetric Learning Asymmetric Nonlearning DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS 30 5.! CONCLUSION Standard form contracts include terms that may benefit consumers and generate costs for the firm in ways that are not perfectly predictable at the outset. Adopting a contract term is often akin to experimentation: the firm may accept the risk of short-term losses in order to learn the net value of the term and take a better-informed decision in the future. Yet, only some terms offer an opportunity to learn and may do so in different ways. We have introduced a distinction between two main categories of terms: symmetric- learning terms are terms that offer symmetric opportunities to learn experientally to firms that adopt them and to firms that do not adopt them; asymmetric-learning terms are those that offer an opportunity to learn either to adopting firms or to non-adopting firms, but not to both. Exploiting differences in the way firms learn from their contractual choices, we have built a theory of experiential learning in standard form contracts. The theory predicts that firms will be more likely to revise terms that offer an opportunity to learn and might fail to revise terms that do not offer such an opportunity. Through this lens, we have examined and classified the terms included in the End User Software License Agreements (EULAs) by a sample of 264 firms across 114 different software markets in 2003 and in 2010. We found that learning opportunities are a determinant of change, overcoming the stickiness of defaults. When such opportunities are absent, terms may survive long enough to appear obsolete and out of touch with the rest of the contract. The analysis we present in this article opens, we hope, interesting avenues for further theoretical and empirical inquiry. To our knowledge, we are the first to identify the learning modalities of different terms and to draw conclusions for contractual choices. Yet, we use a rather rigid, binary classification that does not allow us to distinguish modalities that imply more or less learning. Further research could provide interesting insights into the learning potential of different terms: which terms allow firms to learn the most? Learning also occurs through different channels, as we have emphasized. Understanding how these interact, as well as how new technology affects the way in which firms learn, are important questions that are the subject of future work. Our analysis focuses on learning from direct experience and we have stressed the firm’s behavior in response to information about the costs of offering certain clauses.24 In general, such learning is beneficial because it allows the firm to offer terms that maximize the value of consumer contracts. This observation speaks against the stickiness of default terms: defaults should not be sticky because stickiness distorts the process of learning and prevents firms from opting out of a default terms in cases in which this choice would otherwise be optimal. From a normative viewpoint, the law should make contractual choices as neutral as possible as leveraging on the attractiveness of default provisions comes with a possibly high cost. 24 We recognize that firms may also experiment ways in which they could exploit consumers. There is a large literature about this and similar problems and we do not examine it here. DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS 31 5.1.! Tables Table 2.! Company, Product, Market, and Contract Characteristics Obs Mean SD Min Median Max Panel A. Company Characteristics Revenue 2003 ($000) 259 287,499 2,490,751 30 1700 36,800,000 Revenue 2010 ($000) 259 539,091 4,225,384 90 2200 60,400,000 Change Revenue ($) 254 256,679 1,917,968 -723,200 111.5 23,600,000 Change Revenue (%) 254 226 627 -90 24.08 5000 Public 2003 264 0.11 0.32 0 0 1 Public 2010 264 0.14 0.35 0 0 1 Age 2003 (Yrs) 264 13.62 8.01 0 13 68 Age 2010 (Yrs) 264 20.62 8.01 7 20 75 Lawyers 118 0.74 0.44 0 1 1 Pro-Consumer State 264 0.32 0.61 -1 0 1 Panel B. Product and Market Characteristics Trial 2003 264 0.73 0.45 0 1 1 Trial 2010 264 0.77 0.42 0 1 1 Median Price 2003 ($) 264 812 1,310 14.99 360 12,000 Median Price 2010 ($) 256 841 1,686 8.99 350 20,995 Consumer Product 264 0.36 0.48 0 0 1 Multi-User License 264 0.08 0.28 0 0 1 Developer License 264 0.08 0.27 0 0 1 H-H Index 236 0.37 0.24 .065 .30 1 Panel C. Contract Characteristics Any Terms Changed 264 0.39 0.49 0 0 1 Number of Words 2003 264 1,517 1,365 33 1,152 8,406 Number of Words 2010 262 1,938 2,077 106 1,354 13,416 DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS 32 Table 3.! EULA Terms and Bias: 2003 vs. 2010 EULA terms are classified into 32 common terms that allocate rights and risks between buyers and sellers across seven categories of related terms, according to the degree the terms either match the default rules of UCC Article 2 (Adoption of Default = 0) or deviate from them (Opt-out= 1). “Learning Category” refers to the type and mode that allows sellers to learn from a term. Terms allow for symmetric learning, denoted S, when learning occurs or not regardless of the mode of the term. Some terms allow for asymmetric learning, allowing sellers to learn as long as the mode adopted enables learning. Terms that enable learning when the seller adopts the default rule but not otherwise are denoted A (D) (i.e., asymmetric learning by adopting the default). Terms that enable learning when the seller opts out of the default are denoted A (O) (i.e., asymmetric learning by opting out of the default). The table reports the mean opt-out of UCC Article 2 default in 2003 and 2010, as well as the mean change and statistical significance. * p < 0.10, ** p < 0.05, *** p < 0.01. Learning Category Category and Term Adoption of Default=0 Opt-out=1 Mean 2010 (SD) Mean 2003 (SD) Mean Change (SE) S Acceptance Does license alert consumer that product can be returned if she declines terms? 1 = yes 0 = no 0.458 (0.499) 0.470 (0.500) 0.011 (0.022) Modification and Termination 0.227 (0.539) 0.167 (0.439) 0.061*** (0.021) S Are license’s terms subject to change? 0 = no 1 = yes 0.106 (0.309) 0.076 (0.265) 0.030** (0.012) S Does license allow licensor to disable the software remotely if licensee breaches any EULA terms, according to licensor? 0 = no 1 = yes 0.121 (0.327) 0.091 (0.288) 0.030** (0.013) Scope 1.792 (1.169) 1.659 (1.162) 0.133*** (0.046) S Does definition of “licensed software” include regular updates such as enhancements, versions, releases, etc.? 1 = yes 0 = no or no mention 0.170 (0.377) 0.136 (0.344) 0.034** (0.015) S Can licensee alter/modify the program? 0 = yes or no mention 1 = no 0.640 (0.481) 0.598 (0.491) 0.042*** (0.015) A (D) Can licensee create derivative works? 0 = largely unrestricted or no mention 0.379 0.352 0.027* DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS 35 1 = cap on damages less than or equal to purchase price A (D) Is there an indemnification term? 0 = no, no mention, or twoway indemnification 1 = indemnification by licensee 0.170 (0.377) 0.152 (0.359) 0.019 (0.015) A (O) Maintenance and Support Does base price include M&S for 31 days or more? 1 = yes 0 = no or no mention 0.667 (0.472) 0.663 (0.474) 0.004 (0.014) Conflict Resolution 0.341 (0.513) 0.284 (0.476) 0.057*** (0.019) A (D) Forum specified? 0 = court, choice of licensee, or no mention 1 = specific court or mandatory arbitration 0.322 (0.468) 0.273 (0.446) 0.049*** (0.017) S Law specified? 0 = same as forum or no mention 1 = yes and different from forum 0.011 (0.106) 0.008 (0.087) 0.004 (0.004) S Who pays licensor’s attorney fees? 0 = paid by losing party or no mention 1 = paid by licensee 0.008 (0.087) 0.004 (0.062) 0.004 (0.004) Third Parties 0.216 (0.574) 0.098 (0.346) 0.117*** (0.028) S Does license require licensee agree to third party licenses or terms? 0 = no or no mention 1 = yes 0.121 (0.327) 0.064 (0.246) 0.057*** (0.015) A (D) Does license disclaim licensor’s liability for any included third party software? 0 = no or no mention 1 = yes 0.080 (0.271) 0.034 (0.182) 0.045*** (0.015) S Does license allow licensor or third parties to install additional software? 0 = no or no mention 1 = yes 0.015 (0.122) 0.000 (0.000) 0.015** (0.008) S Consumer Protection Does license inform licensee of statutory rights? 1= yes, contract informs consumer about state law rights they may have 0= no or no mention 0.473 (0.500) 0.417 (0.494) 0.057*** (0.017) Total Mean Change 0.583*** (0.128) DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS 36 Table 4.! Learning and Changing Terms Fraction of terms that change between 2003 and 2010 depending on whether their 2003 values are at the default or, for asymmetric terms, at the learning value. In Panel A, for example, 29.4% of terms were at opt-out values in both 2003 and 2010 and 1.4% were at a opt-out value in 2003 and changed to a default value by 2010. The probability of a change for a term that was at a opt-out value in 2003 is 0.045 (0.014/0.308), while the probability of a change for a term that was at the default in 2003 is 0.056 (0.039/0.692), which is a statistically significant difference of - 0.011. Asymmetric terms can also be at a learning or nonlearning value. * p < 0.10, ** p < 0.05, *** p < 0.01. Panel A. All Terms (32 terms; 8,448 EULA-term observations) 2010 term (Fractions) opt-out default total 2003 term opt-out 0.294 0.014 0.308 Prob(change | 2003 at opt-out) 0.045 default 0.039 0.653 0.692 Prob(change | 2003 at default) 0.056 total 0.333 0.667 1 difference -0.011** Panel B. Symmetric-learning terms (15 terms; 3,696 EULA-term observations) 2010 term opt-out default total 2003 term opt-out 0.238 0.013 0.251 Prob(change | 2003 at opt-out) 0.052 default 0.039 0.711 0.750 Prob(change | 2003 at default) 0.052 DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS 37 total 0.277 0.724 1 difference 0 Panel C. Asymmetric-learning terms (17 terms; 4,752 policy-term observations) 2010 term opt-out default total 2003 term opt-out 0.344 0.015 0.359 Prob(change | 2003 at opt-out) 0.042 default 0.039 0.603 0.642 Prob(change | 2003 at default) 0.061 total 0.383 0.618 1 difference -0.019** 2010 term learning nonlearning total 2003 term learning 0.461 0.036 0.497 Prob(change | 2003 at learning) 0.072 nonlearning 0.017 0.485 0.502 Prob(change | 2003 at nonlearning) 0.034 total 0.478 0.521 1 difference 0.038*** DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS 40 Table 6.! Asymmetric Learning by Default vs. Opt-out Rate of learning values chosen for asymmetric terms, where asymmetric terms are broken down into those where learning is by adoption of the default rules of UCC and those where learning is by opting-out of such default rules. Panel A. Asymmetric-learning terms -- Learning from Defaults (12 terms; 3,168 EULA-term observations) 2010 term learning nonlearning total 2003 term learning 0.555 0.044 0.599 Prob(change | 2003 at learning) 0.073 nonlearning 0.013 0.388 0.401 Prob(change | 2003 at nonlearning) 0.032 total 0.568 0.432 1 difference 0.041*** Panel B. Asymmetric-learning terms -- Learning from Opt-out (5 terms; 1,320 EULA-term observations) 2010 term learning nonlearning total 2003 term learning 0.237 0.018 0.255 Prob(change | 2003 at learning) 0.071 nonlearning 0.026 0.719 0.745 Prob(change | 2003 at nonlearning) 0.035 total 0.263 0.737 1 difference 0.036*** DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS
41 Table 7.! Asymmetric Learning by Default vs. Opt-out Rate of learning values chosen for asymmetric terms, where asymmetric terms are broken down into those where learning is by adoption of the default rules of UCC and those where learning is by opting-out of such default rules. Panel A. Asymmetric-learning terms -- Learning from Defaults (12 terms; 3,168 EULA-term observations) 2010 term learning nonlearning total 2003 term learning 0.555 0.044 0.599 Prob(change | 2003 at learning) 0.073 nonlearning 0.013 0.388 0.401 Prob(change | 2003 at nonlearning) 0.032 total 0.568 0.432 1 difference 0.041*** Panel B. Asymmetric-learning terms -- Learning from Opt-out (5 terms; 1,320 EULA-term observations) 2010 term learning nonlearning total 2003 term learning 0.237 0.018 0.255 Prob(change | 2003 at learning) 0.071 nonlearning 0.026 0.719 0.745 Prob(change | 2003 at nonlearning) 0.035 total 0.263 0.737 1 difference 0.036*** DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS
42 5.2.! Term Classification EULA terms are classified into 32 common terms that allocate rights and risks between buyers and sellers across seven categories of related terms, according to the degree the terms either match the default rules of UCC Article 2 (Adoption of Default = 0) or deviate from them (Opt-out= 1). “Learning Category” refers to the type and mode that allows sellers to learn from a term. Terms allow for symmetric learning, denoted S, when learning occurs or not regardless of the mode of the term. Some terms allow for asymmetric learning, allowing sellers to learn as long as the mode adopted enables learning. Terms that enable learning when the seller adopts the default rule but not otherwise are denoted A (D) (i.e., asymmetric learning by adopting the default). Terms that enable learning when the seller opts out of the default are denoted A (O) (i.e., asymmetric learning by opting out of the default). For each term, the Table reports the rationale support a particular experiential learning classification, as noted in the last column, “Classification Rationale.” Term # Learning Category Term (t) Classification Rationale Learning (0=no; 1=yes) Acceptance x1 S Does license alert consumer that product can be returned if she declines terms? 1=yes; 0=no Notice term giving pure information to the consumer. Feedback about the value of such term is unlikely to arise from direct experience. 0 Modification and Termination x2 S Are license’s terms subject to change? 0=no; 1=yes Term altering the process of contractual modification. Feedback about the value of such term is unlikely to arise from experience. 0 x3 S Does license allow licensor to disable the software if licensee breaches any EULA terms, according to licensor? 0=no; -1=yes Clause makes enforcement of the contract easier. Feedback, through various means, can occur in either case. 1 Scope x4 S Does definition of "licensed software" include updates, enhancements, versions, releases, patches, etc.? 1=yes;0=no mention/no Feature likely linked to consumer preferences and market characteristics. Feedback occurs in either case, though not necessarily from experience. 0 x5 S Can licensee alter/modify the program? 0=yes or no mention; -=no Feature likely linked to consumer preferences and market characteristics. Feedback occurs in either mode, though not necessarily from experience. 0 DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS
45 x19 A (D) Who bears the risk of loss? 0=licensor, for losses caused by factors under licensor’s control, or no mention; 1=licensee Seller learns exposure to liability only if bears the loss. 1 if t = 0 x20 A (D) Who bears the performance risk? 0=licensor, for causes under licensor's control, or no mention, or licensee, for uses expressly forbidden by licensor; 1=licensee (language "licensee assumes responsibility of choice of product and functions, etc.) Seller learns exposure to liability only if bears the loss. 1 if t = 0 x21 A (D) Disclaims incidental, consequential and special damages? 0=no or no mention; 1=yes Seller learns exposure to liability only if there is no disclaimer. 1 if t = 0 x22 A (D) Are damages waived under all theories of liability (contract, tort, strict liability)? 0=no; 1=yes Seller learns exposure to liability only if there is no waiver. 1 if t = 0 x23 A (D) What is the limitation on damages? 0=no mention or cap on damages greater than purchase price; 1=cap on damages less than or equal to purchase price Seller learns exposure to liability only if there is no limitation. 1 if t = 0 x24 A (D) Is there an indemnification clause? 0=no, no mention, or two-way indemnification; 1=indemnification by licensee Sellers from exposure by being liable for any infringement. 1 if t = 0 Maintenance and Support x25 A (O) Does base price include M&S for 31 days or more?1=yes; 0=no or no mention Seller learns only if M&S included. 1 if t = 1 Conflict Resolution x26 A (D) Forum specified? 0=choice of licensee or no mention; 1=specific court or mandatory arbitration Seller learns risks of non-specified forum only if no choice of forum is made. 1 if t = 0 DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS
46 x27 S Law specified? 0=same as forum or no mention; 1=yes and different from forum Sellers receives feedback in either mode. 1 x28 S Who pays licensor’s attorney fees? 0= paid by losing party or no mention; 1=paid by licensee If there is litigation, seller learns anyway the costs. 1 Third Parties x29 S Does license require licensee agree to third party licenses or terms? 0=no; 1=yes Feature likely linked to consumer preferences and market characteristics. Feedback occurs in either case, though not necessarily from experience. 0 x30 A (D) Does license disclaim licensor’s liability for any included third party software? 0=no - 1=yes Seller learns exposure to liability only if there is no disclaimer. 1 if t = 0 x31 S Does license allow licensor or third parties to install additional software? 0=no; 1=yes Feature likely linked to consumer preferences and market characteristics. Feedback occurs in either case, though not necessarily from experience. 0 Consumer Protection x32 S Does license inform licensee of statutory rights? 0=no; 1=yes Pure information term. Feedback occurs in either case, though not necessarily from experience. 0 DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS
47 6.! APPENDIX Proof of Proposition 1. The proposition implies that, if λD > λO, the default is chosen more often at time 0 as the difference λD – λO increases. Since F(p) increases monotonically in p, this is the case if p* increases in λD and decreases in λO. Analogously, if λD < λO the opt out is chosen more often at time 0 as the difference λO – λD increases; this is the case if 1 – F(p*) decreases in λD and increases in λO, which occurs, again, if p* increases in λD and decreases in λO. Thus, it is enough to show that !" ∗ !$% > 0 and !" ∗ !$( < 0. We have: *+∗ *,- = / 01 − 3 − 4 1 + / − /λ8 3 − 09 + λ84 1 + / 01 − 09 − λ- / 01 − 3 − 4 − λ8 / 3 − 09 − 4 : The latter expression is positive if s < w(cH – v), that is, if switching costs are not so high as to prevent switching at time 1, which follows from Assumption 1. Similarly, we have: *+∗ *,8 = − / 3 − 09 − 4 1 + / − /λ- 01 − 3 + λ-4 1 + / 01 − 09 − λ- / 01 − 3 − 4 − λ8 / 3 − 09 − 4 : which is negative if s < w(v – cL) as we assume in Assumption 1. Q.E.D. Proof of Proposition 2. The proposition implies that if λD > λO, there are more switches away from the default at time 1 as the difference λD – λO increases; that is, it implies that Δ(p*) increases as the difference λD – λO increases. Note that Δ(p*) increases in p*. Proposition 1 shows that as the difference λD – λO increases, then +∗ increases, which implies that Δ(p*) also increases. Conversely, the proposition also implies that if λD < λO, there are more switches away from the opt out at time 1 as the difference λD – λO increases; that is, it implies that Ω(p*) increases as the difference λD – λO increases. Note that Ω(p*) decreases in p*. Proposition 1 shows that as the difference λD – λO increases, then p* decreases, which implies that Ω(p*) increases. Q.E.D. Proof of Proposition 3. Since by Assumption 1 we can subsume k under v, it is sufficient to examine the following derivative: *+∗ *3 = 01 − 09 1 + 2 − λ- − λ8 / + 1 − λ- 1 − λ8 /: + λ- + λ8 + λ- 1 − λ8 + λ8 1 − λ- / 4 01 − 09 + λ- + λ8 4 + 01 1 − λ- − 09 1 − λ8 + λ- − λ8 3 /: : Which is positive. Q.E.D. Proof of Proposition 4. We need to consider the following derivative: DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS
50 REFERENCES George A. Akerlof, The Market for ‘Lemons’: Quality Uncertainty and the Market Mechanism, 84 Q. J. OF ECON. 488 (1970). Ian Ayres, Optional Law: The Structure of Legal Entitlements (2005). Douglas G. Baird & Edward R. Morrison, Bankruptcy Decision Making, 17 J.L. ECON. & ORG. 356 (2001) Oren Bar-Gill, SEDUCTION BY CONTRACT: LAW, ECONOMICS, AND PSYCHOLOGY IN CONSUMER MARKETS (2012). Omri Ben-Shahar & John A.E. Pottow, On the Stickiness of Default Rules, 33 FLA. ST. U. L. REV. 651, 655–60 (2006) Lisa Bernstein, Social Norms and Default Rules Analysis, 3 S. CAL. INTERDISC. L.J. 59 (1993) Patrick Bolton and Christopher Harris, Strategic Experimentation, 67 ECONOMETRICA 349 (1999) Stephen J. Choi & Mitu Gulati, Innovation in Boilerplate Contracts: An Empirical Examination of Sovereign Bonds, 53 EMORY L.J. 929 (2004) Stephen J. Choi & G. Mitu Gulati, An Empirical Study of Securities Disclosure Practice, 80 TUL. L. REV. 1023, 1062–66 (2006) Stephen J. Choi, Mitu Gulati & Eric A. Posner, The Dynamics of Contract Evolution, 88 N.Y.U. L. REV. 1 (2013) Stephen J. Choi, G. Mitu Gulati & Eric A. Posner, Pricing Terms in Sovereign Debt Contracts: A Greek Case Study with Implications for the European Crisis Resolution Mechanism, 6 CAPITAL MARKETS L.J. 163 (2011). John C. Coates, IV, Why Have M&A Contracts Grown? Evidence from Twenty Years of Deals (Harvard Law School John M. Olin Center Discussion Paper No. 889, European Corporate Governance Institute (ECGI) - Law Working Paper No. 333/2016 2017). Christopher A. Cotropia, Describing Patents as Real Options, 34 J. CORP. L. 1127 (2009) Kevin E. Davis, The Role of Nonprofits in the Production of Boilerplate, 104 MICH. L. REV. 1075, 1086 (2006) Stefano Della Vigna & Matthew Gentzkow, Uniform Pricing in US Retail Chains, Working Paper Lee Anne Fennell, Revealing Options, 118 HARV. L. REV. 1399 (2005) DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS
51 Clayton P. Gillette, Lock-In Effects in Law and Norms, 78 B.U. L. REV. 813, 819 (1998) Charles J. Goetz & Robert E. Scott, The Limits of Expanded Choice: An Analysis of the Interactions Between Express and Implied Contract Terms, 73 CALIF. L. REV. 261, 286, 289–305 (1985) Mitu Gulati & Robert E. Scott, The Three and a Half Minute Transaction: Boilerplate and the Limits of Contract Design 33–44 (2013) Mitu Gulati & Stephen J. Choi, Innovation in Boilerplate Contracts: An Empirical Examination of Sovereign Bonds, 53 Emory Law Journal 930-996 (2004) Joseph A. Grundfest & Peter H. Huang, The Unexpected Value of Litigation: A Real Options Perspective, 58 STAN. L. REV. 1267 (2006) Jacob E. Gersen & Eric A. Posner, Timing Rules and Legal Institutions, 121 HARV. L. REV. 543, 544-46 (2007) Xinyu Hua & Kathryn E. Spier, Product Safety, Contracts, and Liability, Working Paper David Hoffman, Relational Contracts of Adhesion, 85 U. Chi. L. Rev. (forthcoming 2018). Jason Scott Johnston, Strategic Bargaining and the Economic Theory of Contract Default Rules, 100 YALE L.J. 615 (1990) Marcel Kahan & Michael Klausner, Standardization and Innovation in Corporate Contracting (or “The Economics of Boilerplate”), 83 VA. L. REV. 713 (1997) Kahneman, D., J. Knetsch, and R. Thaler, Anomalies: The Endowment Effect, Loss Aversion, and Status Quo Bias. The Journal of Economic Perspectives 5(1), 193–206 (1991). Godfrey Keller, Sven Rady, & Martin Cripps, Strategic Experimentation with Exponential Bandits, 73 ECONOMETRICA 39 (2005) for a tractable model of experimentation. Michael Klausner, Corporations, Corporate Law, and Networks of Contracts, 81 VA. L. REV. 757 (1995) Korobkin, R. Status Quo Bias and Contract Default Rules. Cornell L. Rev., 83, p.608 (1997) T.L. Lai & Herbert Robbins, Asymptotically Efficient Adaptive Allocation Rules, 6 ADVANCES IN APPLIED MATHEMATICS 4 (1985). Stewart Macaulay, Private Legislation and the Duty to Read—Business by IBM Machine, the Law of Contracts and Credit Cards, 19 VAND. L. REV. 1051 (1966) DARI-MATTIACCI & MAROTTA-WURGLER — LEARNING IN STANDARD FORM CONTRACTS
52 Isabel Marcin and Andreas Nicklisch, Testing the Endowment Effect for Default Rules (2014). MPI Collective Goods Preprint, No. 2014/1. (2014) Florencia Marotta-Wurgler, What’s in a Standard Form Contract? An Empirical Analysis of Software License Agreements, 4 J. EMPIRICAL LEGAL STUD. 677 (2007). Florencia Marotta-Wurgler, Competition and the Quality of Standard Form Contracts, 5 J. EMPIRICAL LEGAL STUD. 447, 457–67 (2008) Florencia Marotta-Wurgler and Daniel Svirsky, Do FTC Privacy Enforcement Actions Matter? Compliance Before and After US-EU Safe Harbor Agreement (NYU Working Paper, 2018). Florencia Marotta-Wurgler and Robert Taylor, Set in Stone? Change and Innovation in Consumer Standard Form Contracts, 88(1) N.Y.U. L. REV. 240 (2013) Robert McDonald & Daniel Siegel, The Value of Waiting to Invest, 101 Q.J. ECON. 707 (1986) ( George Priest, A Theory of the Consumer Product Warranty, 90 YALE L.J. 1297 (1981). Michael Rothschild & Joseph E. Stiglitz, Equilibrium in Insurance Markets: An Essay on the Economics of Imperfect Information, 90 Q. J. OF ECON. 629 (1976). Peter B. Rutledge & Christopher R. Drahozal, Sticky” Arbitration Clauses? The Use of Arbitration Clauses After Concepcion and Amex, 67 VAND. L. REV. 955, 987-1001 (2014) Daniel Schwarcz, Reevaluating Standardized Insurance Policies, 78 University of Chicago Law Review 1263 (2011) Daniel Schwarcz, The Role of Courts in the Evolution of Form Contracts: An Insurance Case Study (working paper, 2018) Alan Schwartz & Robert E. Scott, Contract Theory and the Limits of Contract Law, 113 YALE L.J. 541, 598-601 (2003). Alan Schwartz & Robert E. Scott, The Political Economy of Private Legislatures, 143 U. PA. L. REV. 595, 597 (1995) Alan Schwartz & Louis Wilde, Imperfect Information in Markets for Contract Terms: The Examples of Warranties and Security Interests, 69 VIRGINIA L. REV. 1387 (1983). Alan Schwartz & Louis Wilde, Intervening in Markets on the Basis of Imperfect Information: A Legal and Economic Analysis, 127 UNIV. OF PENNSYLVANIA L. REV. 630 (1979). Kathryn E. Spier, Incomplete Contracts and Signaling, 23 RAND J. ECON. 432 (1992)