Walter Lippmann: an institutionalist for our times?

Abstract

Although Lippmann's The Good Society was written to address the crisis in liberal democracies in the 1930s, we argue that it offers a novel and plausible institutionalist account of the productivity slowdown and the increase in the experience of insecurity that has characterised most liberal democracies over the last 20–40 years. Central to Lippmann's account is a Smithian-institutionalist model of growth where property rights have to evolve both to encourage continued levels of risk taking in the face of new uncertainties and also to offset new sources of unequal bargaining power that the very process of growth itself creates. When property rights fossilise and fail to evolve, as in the 1930s and plausibly also now, productivity growth slows down, insecurity rises, and illiberal political creeds prosper. To avoid this, Lippmann's analysis suggests that property rights have to change to re-energise risk taking and to offset the new sources of unequal bargaining power. For example, in current circumstances, new ‘positive’ property rights arising from the development of social insurance might encourage risk taking and new ‘negative’ property rights in personal data might help offset the new sources of unequal bargaining power that have emerged.

Introduction

Walter Lippmann was a 20th century, left-leaning journalist of influence.¹ He could whisper in the ears of Presidents. This began with Woodrow Wilson over entry into WW1 and ended with Lyndon Johnson, when the Vietnam War provoked Lippmann into something more like shouting at the President. He won two Pulitzer prizes for his journalism and wrote several books. One of those books, The Good Society, published in 1937, addressed the crisis in liberal democracy in the 1930s and what needed to be done to revive its fortunes. In this paper, we present and commend the institutional model of growth that is a part of the argument in that book. We commend the model in part because it provides a novel institutional account of what are often taken to be characteristic features of the contemporary conjuncture in many rich countries: the simultaneous slowdown in productivity growth and the increased experience of insecurity – hence the title of the paper.

We are not directly concerned, therefore, with Lippmann's position in the liberal tradition or where he fits in with the specific debates of the 1930s. We are distilling, presenting and commending his institutional model of growth for the present. Nevertheless, we will return at the end to make some comment on his type of liberalism and it helps to begin by putting the book in some historical context. Although now largely forgotten, The Good Society was immediately enthusiastically received, among others, by liberals such as Friedrich Hayek (Jackson, 2012). A year later, following the publication of a French edition of The Good Society, the philosopher Louis Rougier organised the Colloque Walter Lippmann to discuss Lippmann's ideas about the causes of liberalism's decline and the prospects for its revival. The Colloque now has a prominent position in the history of how efforts to revive liberal thought from the late 1930s onwards led, via the formation of the Mont Pelèrin Society in 1947, to the rise of so-called neo-liberalism (Burgin, 2012; Cockett, 1994).² However, while Lippmann and Hayek agreed on the need to revive liberal thought in the face of the rise of collectivism in the interwar years, they differed significantly on the vision of liberalism, and the kinds of policies, that revival would entail. For example, Lippmann viewed liberalism as being consistent with Keynesian demand management and also came to emphasise, to a significantly greater degree than did Hayek, the importance of government intervention to control corporate power. Most importantly for the purposes of this paper, Lippmann advocated new forms of social insurance that would protect people against the uncertainties arising from the process of economic growth, thereby encouraging them to resist the siren calls of the advocates of collectivism.³ Accordingly, and notwithstanding his important role in the history of the rise of the Mont Pelèrin Society and neo-liberalism, Lippmann increasingly distanced himself from Hayek: he declined an invitation to write a Preface to Hayek's Road to Serfdom; and he never attended a meeting of the Mont Pelèrin Society (Burgin, 2012: 85–86; Goodwin, 2014: 255–57).⁴

Lippmann's account of the challenges to liberalism that emerged in the 1930s turns on the failure of liberal institutions, particularly property rights, to adapt to the new sources of uncertainty and unequal bargaining power that had emerged through the very process of growth since the 19th century. Growth slowed down as a result and life was increasingly insecure. The combination of growing insecurity and stagnating living standards was the breeding ground for authoritarianism. We commend the institutional model of growth that underpins this analysis, for three reasons.

First, Lippmann's institutional model of growth is both plausible and in some respects novel. It is built upon an understanding of productivity growth that comes in part from Adam Smith and the division of labour.⁵ For Smith, an extension of the division of labour, the engine of growth, is always something new and a step into the unknown. Lippmann follows Smith, as many economists have since, in arguing, as a consequence, that growth depends on risk taking, broadly understood.

He goes on to argue, in a more novel way, that the extant property rights regime influences how risks are perceived at that time. The risk mediating quality of property rights is part of the way that they structure incentives towards action. In particular, the ‘negative’ aspects of property rights establish what can and cannot be done without possible sanction, either by the State or through the actions of others. In this way, they establish boundaries for some of the uncertainty in decision making: i.e. some of the uncertainty that comes from what other agents may do in any setting. However, property rights, like all rights, also have ‘positive’ aspects (Berlin, [1958] 1969). The ‘positive’ aspect of property rights establishes what a person can expect to have in any situation, through collective institutions like the State, by virtue of being a citizen; and they too influence the perception of risk in any setting. For example, the marginal tax rates and the existence of unemployment, healthcare and other benefits will establish boundaries to the possible downsides that attach to decision making when there is uncertainty. The fact that property rights in both senses affect the perception of risk is important because people are generally averse to risk. Thus, depending on the character of the extant property rights in relation to the risks at that time, they can, but may not, encourage the risk taking that underpins growth.

Finally, and distinctively, Lippmann argues that growth itself creates new risks. For example, each extension of the division of labour adds complexity to the economy, resulting in potentially greater unpredictability and insecurity. This completes the first strand of his argument as to why property rights need to evolve. The process of growth builds in, as it were, a sunset clause for any set of property rights. The property rights that now encourage risk taking and fuel productivity growth are bound to create new sorts of risk and insecurity and, without a change in those property rights, risk taking will slow in the presence of these new risks and with it productivity growth. In short, Lippmann provides a distinctive institutional model of growth (different, e.g. from the ‘new institutional economics’) whereby property rights have to adjust dynamically to the endogeneity of people's appetite for taking risks.

In Lippmann's analysis, then, the liberal rights that are characteristic of Western societies, and particularly their property rights, have to evolve to secure continuing productivity growth. If they do not, growth slows and insecurity rises. The liberal order of rights becomes associated with economic failure and people are attracted to non-liberal ways of organising society that promise greater security in one way or another (Goodwin, 1995: 335). This is what had happened in the 1930s according to Lippmann. The idea of liberty had become fossilised in a particular 19th century laissez-faire incarnation, most closely associated by Lippmann with the work of Herbert Spencer. Rights had stopped evolving in ways that facilitated growth and liberalism was no longer delivering economically. The ‘liberals failed to develop the promise of liberalism’, Lippmann (1937: xiii) maintained, and so ‘ceased to … command the interest of the people’.⁶ To recover its political influence, liberalism would have to reimagine what liberty means so that it is tailored to fit a new set of circumstances. Critical in this, but not exclusively so for Lippmann, was the redefinition of rights to provide for greater security in a world that is otherwise necessarily becoming more uncertain. This was how the siren voices of the collectivists in the 1930s offering security in other ways were to be out-manoeuvred.

Property rights also need to evolve for another reason in Lippmann's analysis. In this respect, he draws more closely on the conventional economic theory of his time to argue that the absence of competition in markets causes inefficiency. The generic source of an absence of competition is unequal bargaining power and the origins of this are always to be found in the prevailing property rights regime. Monopolists, for example, always have some right that protects their monopoly position from erosion (e.g. through a patent) and/or they have a right to exploit such a monopoly position (e.g. through being able to set any price). Again, Lippmann's argument turns on the way that growth through the division of labour transforms the economy into some new shape or form. In the past, the property rights regime may have prevented the exercise of unequal bargaining power, but the transformation caused by growth is liable to create new sources of such power and so the rights regime has again to evolve to prevent this proving a drag on prosperity.

We set out this model and its distinctive features in more detail in the next section.

Our second reason for commending this model of growth is that it passed a Popper-like test. In particular, in the later parts of The Good Society, Lippmann uses his institutional theory of growth to sketch the kinds of changes to rights that would be necessary if growth and security were to be restored and so provide the conditions for the revival of liberalism in his time. Loosely speaking, one could say that Lippmann used his institutional model to make some conjectures regarding what would be necessary for the revival of liberalism. In many respects, these conjectures anticipated the actual growth of the welfare state and the human rights revolution that accompanied the post war revival of liberalism. Thus, his institutional theory has ‘successfully’ passed, in a loose sense, a Popperian-lstyle test with respect to the generation of conjectures regarding how to revive liberalism that were not subsequently falsified.⁷ We develop this claim in the third section.

Our final reason comes in the fourth section, where we argue that there is prima facie evidence to support the contemporary operation in many of today's rich countries of the mechanisms that Lippmann identifies as responsible for stagnating living standards and increased insecurity.⁸ In other words, Lippmann's model of growth can be used to help explain the contemporary productivity growth slowdown and rise in insecurity that has been experienced by many of today's rich countries; and it thereby sets a policy agenda for how to restore growth and check insecurity in those countries. The contemporary challenge to liberalism from populism is also often associated with this stagnation in living standards and with rising insecurity (see Margalit, 2019).⁹ Thus, the application of his theory of growth might also contribute in this way, for those who are concerned to do so, to the contemporary revival of liberalism's fortunes.

In the final section, we conclude by reflecting on the changes in property rights that the application of Lippmann's model of growth to the present might suggest would be necessary to combat the contemporary experience of stagnant incomes and the perception of heightened insecurity. We focus on the ways that the ‘positive’ aspects to property rights might be developed with new forms of social insurance so as to combat the unpredictability and insecurity that has arisen from greater complexity of the economy.

Productivity growth through the division of labour: the evolving need for risk taking and the avoidance of unequal bargaining power

The argument in outline

Following Adam Smith's analysis in The Wealth of Nations, Lippmann's engine of growth is the division of labour (1937: 161–65, 177, 180). As Lippmann put it, ‘Adam Smith discerned the basic truth that the new industrial technic consists of the division of labour regulated in markets … He saw that the increasing division of labour was the essential revolution in modern times.’ In Smith's model, the division of labour fuels productivity growth for reasons illustrated by his famous pin example. And for Lippmann, like Smith, the division of labour is not just generating productivity growth in the factory; it applies equally to the production of knowledge and technological change. It is the ubiquitous motor of improvement. This makes what determines the extent of the division of labour a key variable in the analysis of growth. Smith argues that it depends on the size of the market and this, in turn, increases with the growth of productivity that comes from the division of labour so as to produce a potentially self-reinforcing process involving the size of the market, the division of labour and productivity growth.

According to Lippmann, this dynamic process was facilitated by the gradual replacement of ‘the political technology of the pre-scientific age’ – whereby resources were allocated within small, self-sufficient communities according to the dictates of custom and tradition – by a liberal regime of freedom under the law that afforded individuals the opportunity and the incentive to act as they saw fit and to break with established routines (Lippmann, 1937: 9; also see pp. 7–19, 165–68, 204–09, 241–46). In particular, like Smith, Lippmann saw clearly that to get and stay on such a self-reinforcing path, there has to be a mechanism for coordinating the economic activities of people who increasingly specialise in what they do, and are therefore increasingly dependent on each other, but who also enjoy growing freedom to act as they wish (Lippmann, 1937: 174–75, 180–82, 237). This is where the market and property rights play a crucial role in Smith's and Lippmann's analysis of growth. Their role in coordinating the specialised activities that come from a division of labour is sketched by Smith and Lippmann brings some distinctively Austrian arguments developed during the 1930s debate over the possibility of central planning to his version of this argument. They need no rehearsing here (Lippmann, 1937: 11–12, 24–35, 93–101, 168–76, 192–94, 229–30, 362–64).

The role of property rights, however, is given a particular emphasis by Lippmann that is significant for his (and our) broader argument. Property rights are important for the operation of markets because, self-evidently, they define what is being exchanged. People need to know what they are exchanging before they can decide whether to or not. While this is a natural way to think of the function of property rights, Lippmann gives a slight twist by expressing what they do in terms of providing ‘security’ (1937: 241–3). We might say today that property rights can mitigate risks. Why would anyone, for example, purchase something in a shop if there was a real risk of being mugged on the way home? What property rights do in such cases is take a real risk out of an exchange by placing limits on the outcomes associated with purchasing a commodity. The limits will depend on how extensively property rights are defined, but minimal property rights will supply an expectation that the object, once purchased, will remain in the possession of the purchaser.

This is a subtle twist and Lippmann, as in his account of why property rights evolve, is following Smith's analysis of their evolution in Book III of The Wealth of Nations. Their development facilitates material improvement, on Smith's account, because it brings greater security to economic decision making. Actions have more predictable outcomes than they otherwise would.

Order and good government, and along with them the liberty and security of individuals, were in this manner established in cities at a time when the occupiers of land in the country, were exposed to every sort of violence. But men in this defenceless state naturally content themselves with their necessary subsistence; because to acquire more might only tempt the injustice of their oppressors. On the contrary, when they are secure of enjoying the fruits of their industry, they naturally exert it to better their condition…. (Smith, [1776] 1904: 377; Bk. III. Ch. III)

The same point is more usually expressed today as the requirement of ‘incentive compatibility’ for efficiency: that is, an alignment between the person who makes the effort and the person who receives the reward from that effort. This is, of course, an accurate description of Smith's point. But it also glosses over how the development of property rights, in effort in this instance, removes an uncertainty – deriving from being ‘exposed to every source of violence’. It is not that the predations of the oppressors are guaranteed in the countryside, it is that oppressors ‘might be tempted’ to predation. This observation does not change the ‘incentive compatibility’ point because it is not every change that might yield greater certainty of outcome which would also produce material improvement. Nevertheless, it is important for Lippmann that the elaboration of property rights removes a risk from engaging in all kinds of improvements. Investing in improvements is a risky business, people do not like risk, and the mitigation of those risks will therefore encourage those improvements.

The risk-mitigating attribute of rights is a more general feature of property rights. For example, in Hobbes's original argument for the State to hold the monopoly of force in a society, people thereby obtained a new right to expect that others could not use force against them. The risks associated with life were reduced, so people could concentrate on productive activities rather than trying to insure themselves through building defences and arming themselves against the risk of pillage by others. Likewise, in the now familiar problem of common pool resources, free riding can create inefficiency, as when the production and increase of greenhouse gases causes climate change. However, when ownership rights are created in such resources, the risk of free riding is reduced because an individual's use of the resource now requires the agreement of its owners. Again, rights can be productive because they reduce risks.

Rights are also potentially productive for Lippmann because they help remove distortions that are a source of inefficiency. For example, rights to equal pay for equal work are productive because they avoid the distortions or grants of privilege that arise when, say, women or ethnic minorities are treated differently on grounds of gender and race and so become confined to activities where they are less productive than they might otherwise be (Lippmann, 1937: 348–51; cf. Boettke and Candela, 2017: 150). Similarly, a rule circumscribing the predatory use of monopoly power gives rights to those who lack such power, attenuating the deadweight losses associated with monopoly.

These examples show that while the creation of rights can be productive in risk mitigating and other ways, this will depend on the character of the rights in relation to whatever is the potential source of productivity at the time. It is not an intrinsic property of any set of rights per se. Indeed, this is why Lippmann argues that rights need to evolve. They need to change in specific ways to address the novel risks and new sources of unequal bargaining power that are bound to arise in the process of growth if that process is to continue: productivity growth both requires risk taking and also generates new kinds of risk and new sources of unequal bargaining power.

Smith, the Austrians, and some other economists have always argued that risk taking underpins growth,¹⁰ but the idea that growth itself creates new risks is something that is distinctive in Lippmann's analysis. It is crucial to Lippmann's argument for why rights need to evolve. This is what happened, for instance, in the 19th century when property rights were elaborated to allow for ‘limited liability’ companies (Lippmann, 1937: 13). This was a risk-mitigating innovation in property rights that sustained the momentum of risk-taking behaviour by corporations. And unless rights evolve in this way, people, who are naturally risk averse and face a riskier world as it grows, will increasingly become reluctant to take the kind of risks that keep productivity growing. Prosperity stalls at the same time as the experience of insecurity is increasing. The two are directly linked by the failure of rights to evolve and this combination proves to be the breeding ground, for Lippmann, of the collectivism of the 1930s.

There are two important aspects of this argument that we now consider in more detail: how growth, ceteris paribus, increases risks; and how it embodies a distinctive form of institutionalism.

Why growth tends to increase risks

Lippmann offers several reasons why growth itself creates new risks. The central one is the extension of the division of labour. This makes the economy complex in the sense that it becomes an interconnected system where developments in any one part can come to depend on events in any other part. The more complex the economy becomes through the extension of the division of labour, the more difficult it becomes for anyone operating in one part to predict what might happen in all of these other interconnected and potentially influential parts (Lippmann, [1934] 1992: 49). Indeed, the outcome for any one person may depend on small changes in behaviour by someone, somewhere else in the system of interdependence about which it is becoming increasingly difficult to know. In this way, the growth of productivity through the division of labour is accompanied by a growth in the experience of risk by its inhabitants. In effect, Lippmann is anticipating here a tendency towards a specific form of what sociologists later called, albeit for different reasons, the ‘risk society’: that is, a progressive heightening in the experience of risk in daily life (Beck, 1992). For Hayek, an appreciation of the same point about complexity is the source of his famous knowledge problem (Hayek, [1937] 2014: 72–73).

This built-in tendency towards risk proliferation matters because people are typically risk averse. Thus, if doing something ‘new’ exposes people progressively to more of these background risks as the complexity of an economy increases, and an economy becomes more complex with growth, then, ceteris paribus, growth will progressively erode people's willingness to do the ‘new’ things that are the very source of growth – unless, of course, there is some compensating extension in their rights through, for example, new forms of social insurance that will counter the influence of growing background risks.¹¹

This general mechanism is amplified by a political one for Lippmann. Any fresh extension in the division of labour always creates new winners and losers and the latter are a possible constituency for a political challenge to liberalism.¹² For while ‘the market does regulate the allocation of capital and labor with some efficiency’, prices ‘are often unreliable indicators of what men who have labor or capital to invest ought to specialise on’ (Lippmann, 1937: 171). Markets therefore operate with ‘a very large margin for error, which in human terms means personal misery’ arising from the way that misallocations of resources are ‘sufficiently violent to wreck many lives before men can readapt themselves’ (Lippmann, 1937: 171–72):

It is easy to understand, therefore, why almost all men felt that they must escape the ruthless dictation of the open market. The collectivist movement in its many manifestations is, I believe, precisely that – a rebellion against the market economy … a demand for protection … Men begin by seeking protection from a mastery of markets. They end by rejecting the whole conception of an economy in which the division of labour is regulated in markets. Instead they adopt the conception of regulation by intelligent authority. (Lippmann, 1937: 172–73; also see pp. 210–12)¹³

The losers, therefore, have to be insured in some degree against their possible losses; otherwise they may be persuaded against liberalism by those in politics offering security through collectivist interventions (Lippmann, 1937: 210–12, 234–38). The inherited system of rights may have bought the agreement of past losers to the changes in production, but it will not always provide similar insurance for the new losers created by the next extension of the division of labour. Hence, the need for the continued revision and updating of the liberal agenda: ‘liberalism is not the rationalization of the status quo but the logic of the social readjustment required by the [ongoing] industrial revolution’ (Lippmann, 1937: 225; also see pp. 212–18, 355–61).¹⁴

A distinctive institutional model of productivity growth?

This model is broadly institutionalist in the sense that it can readily be framed in terms of how the rules guiding action evolve to secure continuing growth of prosperity (or do not). More specifically, Lippmann's approach has much in common with American institutional economics, many of whose leading exponents viewed the compatibility – or lack thereof – between the property rights regime and developments in technology as a key issue (cf. Goodwin, 1995: 322, 335, 2014: 14–16, 42–43).¹⁵ Lippmann expresses his argument most of the time in terms of rights, especially property rights and their evolution, and, in this sense, he focuses on formal ‘rules’.¹⁶ But it is common to treat formal rules and rights as two sides of the same coin, so this is not an important difference. Lippmann is, nevertheless, a very particular kind of institutionalist. What makes for his particularity are three attributes.

First, he is what might be termed a ‘sunset institutionalist’. The institutions that interest him have built-in sunset clauses. This is because the process of growth endows them with this quality: their very success undermines their suitability for future success. They, therefore, become time limited and what broadly explains stagnation are the psychological, social and political sources of inertia that mean societies cling on to the ‘old’ ways of doing things long after their due dates.

Second, although Lippmann focuses on formal rules and their associated property rights, he does not conceive, as is commonly the case, of property rights simply in a negative sense. Instead, just as rights in general can be conceptualised as ‘positive’ as well as ‘negative’ (e.g. see Berlin's famous 1958 essay), Lippmann recognises that the constellation of property rights have ‘positive’ as well as ‘negative’ aspects.¹⁷ The point is this. Property rights determine what a person owns and what they can do with what is owned without possible sanction or counteraction by another. They determine, in other words, an individual's action set and its likely consequences for that individual. Statute and common law contribute to these rights as they establish when and where a person is liable for harms caused by their actions. It is natural to cast this aspect of their rights as ‘negative’ because these laws broadly indicate what a person is free to do. However, these are not the only rules that are relevant to what a person owns, what they can do and with what likely consequence. The rules governing taxation and the State's provision of goods and services, for example, also affect what a person has and what they can expect to happen when they act. These rules in society bestow positive rights; and they are often property rights: for example, a person is entitled to unemployment benefit when unemployed or to have access to medical care when ill, etc.

Third, Lippmann identifies two features of property rights as crucial for productivity growth: their encouragement to risk taking; and the checks they provide on sources of unequal bargaining power. These are the features of the rules that, given the first sunset insight, policy makers need always to follow and be ready to adjust. We develop this last point and the one about ‘positive’ rights when we consider in the Conclusion what this analysis might imply for policy making today.

A test: Lippmann's anticipation of the post WW2 political revival of liberalism

In this section, we suggest a different reason for taking Lippmann's argument seriously now. He used his model to make what, in effect, are predictions or conjectures about what would be required to restore liberalism to its polestar position in politics. Many of these predictions were successful in identifying features of the actual post WW2 revival of liberalism in the democracies of North America and Europe. We offer four exhibits to this effect.

First, there is the development of the redistributive welfare state after WW2 and the acceptance of Keynesian demand management. Lippmann explicitly endorses Keynes's proposal because demand management can reduce macro risks (Lippmann, [1934] 1992: 45–60, 1937: 218–19, 229–30)¹⁸ and he likewise argues, see below, for an expanded welfare state because it helps contain risk.¹⁹

There is no reason whatever why some part of the wealth produced should not be taken by taxation and used to insure and indemnify human beings against their personal losses in the progress of industry … if society as a whole is richer when an industry moves from a place where costs are high to one where they are lower, then some part of that increased wealth can be used to relieve the victims of progress. It can be used to tide them over while they are changing their occupations, to re-educate them for new occupations, to settle them in new places if they have to move … For if it is properly devised, such a system of social insurance would facilitate the necessary technological changes, and reduce the very human resistance which comes from those who now see themselves the appointed victims of progress. No one can blame a man for hating a machine that will place him in the bread line and unfit him for the only job he has learned to do. (Lippmann, 1937: 223–24)

Second, there is the policy shift in North America and Western Europe from the cartels and controlled markets of the inter-war period to the pro-competition market policies of the latter part of the post-WW2 era. Lippmann recommends such a policy because he wants to tackle the unequal bargaining power that can lead to monopoly-like distortions in resource allocation; and in tackling this he is also addressing a particular source of inequality (1937: 220–26). He refers to this as a problem of ‘unearned’ income. It is ‘unearned’ in the sense that it comes from having the power to control supply, raise the price above its competitive level and so transfer some of the benefit of the exchange from the consumer. It serves no useful economic role as it contributes nothing to the expansion of the pie.

Another concern arises from the greater mobility of capital than labour, which makes labour bend to capital. Lippmann wishes to reverse this by changing the rules in ways that will encourage capital to follow labour.

On the whole, the machines must come to the men rather than men to the machines. A civilized life is impossible for nomads who settle nowhere and do not put down deep roots in a particular place. For men who have just arrived and will soon depart tend to be crudely acquisitive. They are transients who have no permanent stake in any community, and there are no ties, other than the cash nexus, between them and their neighbors. They live only in the present, having no ancestral tradition fixed on any place and no care for posterity. The good life finds little encouragement where men do not feel themselves to be links in a chain from the past into the future, where they live from day to day without deep associations and long memories and more than personal hopes. (Lippmann, 1937: 213–14)

Encouraging such mobility will require, in his view, a change in company law. Profits will have to be distributed in dividends: they cannot be used for re-investment because this encourages inertia.

If capital is to achieve the necessary mobility, it must not become entrenched, uneconomically bottled up, in certain favored corporate structures. This is what happens when the managers are permitted to retain the profits, over and above sinking funds and working reserves, and to invest them, without submitting to the test of the competitive capital market. (Lippmann, 1937: 215–16; also see pp. 277–79)

This last line of argument is worth mentioning not for the way that it anticipated an aspect of the post WW2 revival of liberalism (because it did not!) but rather for its contemporary relevance (Waldinger and Schultz, 2023). For instance, the contemporary rise of populism is often associated with the way that the experience of globalisation has undermined the integrity of many local communities, diminishing people's sense of belonging, and leaving them poorer and reliant on moving to find jobs elsewhere (e.g. see Goodhart, 2017, on the origins of the Brexit vote). Lippmann had just such a concern in his sights.

The third exhibit comes from a Google Ngram on the relative frequency of the terms ‘human rights’ and ‘social justice’. This is shown in Figure 1. While references to both grew after WW2, ‘rights’ talk powered ahead of references to ‘social justice’, especially after 1970. This is when the new rights were being established to avoid (inefficiency generating) discrimination around race, gender and sexuality. This switch in emphasis is as Lippmann thought it should be: what matters is the constellation of rights in society and not some pattern of outcomes in the name of social justice or any other desideratum. Any focus on outcomes not only runs up against the knowledge problem that any central authority in a complex economy must face in striving to achieve such goals; it also fails to point politics at what matters, namely the rules that bestow the rights determining how liberty is understood.

T H E prospects of freedom depend very largely upon whether the intellectual leaders of the modern world can recover the intellectual habit of looking for a solution of social problems by the readjustment of private rights rather than by public administration….When contemporary men are confronted with a problem, they no longer inquire whether it can be solved by altering the reciprocal rights and duties of individuals: the preferred solution is almost invariably to invest officials with authority to enforce a solution … Truly conceived, a democracy is not the government of a people by elected representatives exercising the prerogatives of their former lords and masters. It is the government of the people by a common law which defines the reciprocal rights and duties of persons. This common law is defined, applied, and amended by the representatives of the people. (Lippmann, 1937: 266)

Figure 1. Source: Jorda et al (2017).

The fourth and final exhibit follows from the third: it is the rise of the regulatory state.²⁰ If politics has to focus on fine tuning and elaborating the rules of the game and the associated rights people enjoy in an increasingly complex economy and society, then the task of implementing the rules will become increasingly difficult and so have to be devolved to specialist agencies.

… it is obvious that these functions can be performed only by experts using specialized technical procedure. …. Such a function has to be delegated. ….The commissioners may much more properly be looked upon as men entrusted with tentative legislative authority subject to review by the representatives of the people.

Thus the more the legislature delegates its authority to specialized organs of government and the less it endeavors to define the precise law for complex human affairs, the more indispensable is it that the state, through the courts and through the legislature itself, should regard itself as a tribunal to review the conduct of these specialized lawmakers. (Lippmann, 1937: 301–02)

In short, when more complicated rules are applied to an economy and society that is also more complicated, expertise is required. Politicians cannot be directly involved in setting the level of unemployment insurance, the interest rate or the length of a patent, for example, because they cannot possibly know enough to make such judgements. Experts are needed.²¹ This, in turn, gives politics a further objective. It should not just focus on fine tuning the rules; it also has to oversee the actions of the expert bodies that implement them. Indeed, the balance in these activities for politicians will necessarily have to shift in the direction of the latter. The judiciary is, of course, also involved in the evolution of the rules, their interpretation and in an oversight role and so Lippmann is offering what becomes a familiar account of checks and balances or polycentric governance.²²

The contemporary slowdown in productivity growth: a growing reluctance to take risks and new inequalities in bargaining power?

In this section, we develop a final reason for taking Lippmann's institutional model of growth seriously. The model could potentially be helpful in understanding the contemporary crisis of liberal institutions. This is not to say that contemporary crisis is the same as the one of the 1930s that the model was developed to address. They are plainly very different in many respects. Nevertheless, the contemporary crisis is associated with stagnating living standards and the increased perception of economic insecurity just as was the crisis in the 1930s (see Standing, 2011, on the contemporary experience of economic insecurity, and Bergeaud et al., 2016, on the slowdown in productivity growth in most OECD countries since 1975–80; and see Margalit, 2019, and the references therein, for the possible connection between these and rise of populism).²³ The absolute income levels may be higher now and the economic insecurity may not be comparable with that of the 1930s, but the one has been stagnating and the other has been increasing in rich countries recently.²⁴ It is this fact that makes Lippmann's model of growth potentially attractive for the analysis of the present because it compactly ties one of these distinguishing features (the slowdown in growth) to the other (the decreased appetite for risk taking that comes from the experience of economic insecurity). It is a failure of property rights to adjust that heightens economic insecurity which in turn saps the appetite for taking the risks which underpin growth; and so the model directs a policy for the contemporary revival of growth (and possibly the fortunes of liberalism) towards an appropriate adjustment of property rights now. We discuss this possibly useful policy implication in the next section; for now we focus on whether there are reasons for believing, in the first place, that Lippmann's model of growth might help explain the contemporary slowdown in growth in rich countries.

We approach this question in two ways. First, we consider whether there is some prima facie evidence suggesting that the two specific mechanisms responsible for slower productivity growth in Lippmann's model have indeed been in play in run up to the contemporary crisis of liberalism. There is. We then briefly consider the alternative explanations of the slowdown so as to put the Lippmann one in better perspective.

One mechanism in Lippmann's analysis of stagnating living standards is that the experience of greater economic insecurity reduces the willingness to take the risks that spawn productivity growth. There is some evidence of this reduced willingness from around 1980 in the major rich countries in the Jorda et al. (2017) data on the risk premium in these countries. We reproduce this evidence below in Figure 2: the risk premium is measured by the difference between the return on equity and property and the return on risk-free government debt in the major rich countries. The risk premium, on this measurement, is volatile but, while it fell steadily in the first part of the post war period, from about 1980 it began trending upwards (i.e. at around the same time as productivity growth began to slow down).

Figure 2. Ngram: human rights and social justice.

Of course, the risk premium in financial markets is an equilibrium outcome and could arise from two polar sets of circumstances. In one, people progressively shy away from risky actions (the implied conjecture from applying Lippmann's analysis) either because they have become more risk averse or the risky actions have themselves become more uncertain, making them less attractive to anyone with unchanging risk preferences. If people's inclination (demand) to take risk falls for either reason, the relative price for taking risk will rise in equilibrium – as we observe. Alternatively, though, the risk premium could have risen because the returns from risk taking have independently increased (i.e. it is the supply of risk-taking opportunities that has changed and not the demand for them). Such a change in the return to risk taking encourages people with unchanged risk preferences to take on more risks precisely because they allow the risk premium to rise. There is no evidence, however, in support of such an exogenous increase in the return to risk taking in Jorda et al. (2017). Indeed, the return on equity and property remains roughly constant throughout the post war period. Instead, the risk premium rises largely because the return on safe assets has fallen, as would happen if people's preferences for risk had fallen, increasing their demand for safe assets.

The demand side interpretation receives further support because there are other, complementary reasons for supposing that people may have become more reluctant to take risks. The appetite for risk depends, in principle, on people's preference for risk and upon how much risk they perceive as attaching to risky actions. On the former there are reasons for supposing that preferences may have become more risk averse. For example, the greater weight of older people in the demographic profile of most rich countries would likely, ceteris paribus, lead to greater risk aversion on average as the old are typically more risk averse than the young. Likewise, women are apparently typically more risk averse than men (Falk et al., 2018) and so their increased participation in the labour force, particularly in decision-making roles, would likely have the same effect on average risk preferences.²⁵ On the latter, the perception of risk, there is also some evidence pointing to an increase. We reproduce, in Figure 3, a version of the Almelhem et al. (2023) inspired Ngram from John Burn-Murdoch's article in the Financial Times, 5 January 2024. We likewise focus on the balance between words related to progress and improvement versus caution, threat and risk but consider only English language books since 1800. It will be apparent, as in Burn-Murdoch Ngram, that the balance between progress and caution on the one hand and threat and risk on the other moved towards the latter in the last 20–30 years of the 20th century.

Figure 3. Ngram: ‘progress and improvement’ versus ‘caution, threat and risk’.

It should also not be so surprising to find such a growing perception of risks given the plausible rise of so-called ‘background risk’ in recent times. Two recent putative sources of increased ‘background risks’ are noteworthy because they can be directly related to Lippmann's analysis. They are climate change and the developing geo-political uncertainties associated with the rise of China. Geo-political risks, like those associated with the increasing role of China in the world economy, are a good example of the way that productivity growth generates greater complexity in the economy and therefore, ceteris paribus, greater unpredictability. The climate emergency can also be directly related to a failure of the rights regime to evolve. There was no need for rights in the atmosphere during the early stages of productivity growth because the atmosphere can absorb a certain amount of greenhouse gases. It is only after a long period of growth that this threshold level is reached and rights have to be created in the atmosphere to prevent it being breached.

The other Lippmannian mechanism undermining productivity growth centres on an expanding range of inequalities in bargaining power which make markets less efficient. Again, there is evidence pointing to this. For example, part of the increased income inequality in most rich economies since the 1980s is attributable to changes in bargaining power, involving for example increasing product market concentration at the same time as declining union membership (Bajgar et al., 2021). Most recently, product market concentration has been conspicuous in the new media industries, where it has been linked, in a way that again fits with Lippmann's argument about the need to re-imagine the rights regime, to a failure to develop individual rights in personal data.

The alternative explanations of the productivity slowdown in rich countries have typically identified two types of causes: a slowdown in technological change and ‘adverse’ changes in institutions, broadly understood (e.g. see the lists of causes in Gordon, 2012 and Haldane, 2015). The latter contains such things as the inequalities in bargaining power that Lippmann mentions as well as other developments like the possible fall in generalised trust in many countries. In this respect, there is, therefore, overlap between Lippmann and the current literature.

The putative decline in technological change might, in principle, also reveal additional common ground with Lippmann's analysis because the decline in technological change could have occurred, in the manner of Lippmann, when people became increasingly reluctant to take risks. However, there is no mention of this possibility in the list of causes of the slowdown in the recent literature. Instead, this literature typically assumes that nature has become increasingly reluctant to reveal its secrets. Technological change has, as it were, just become more difficult – perhaps because we have already garnered the low hanging in the past.

This technological pessimism may seem plausible on the basis of the likelihood of diminishing marginal returns applying to most things. But, knowledge is not quite like most things and neither is a generalised expectation that there will be diminishing marginal returns in the acquisition of knowledge entirely consistent with the history of science. For example, it is common to distinguish between ‘normal’ science, for which diminishing marginal returns might well apply, and ‘paradigm’ shifts where it need not (Kuhn, 1962). Of course, paradigm shifts may have become less frequent; and there is some evidence to this effect (Youn et al., 2015). The key question, though, for successful policy formation is: why has this occurred? Has it really become more difficult or have we become less inclined to take the risks that paradigm shifting work requires? Lippmann alerts us to this second possibility and it does not feature in the existing literature on the slowdown. This is, in part, why Lippmann's analysis is important.

Conclusion

We have not attempted to make sense of Lippmann's historical location in this paper. He seems to us to belong to the part of the liberal tradition that conceives of liberty arising within a set of rules of the game. The task for liberals, he believes, is to get the rules right so that individuals can flourish through the exercise of their freedoms within those rules. In this respect, he is not unlike Hayek in The Constitution of Liberty.²⁶ It also explains his attraction to Keynes's macroeconomic policy. Both he and Keynes thought this policy would provide the macroeconomic stability within which people are able freely to make the micro-level decisions over what to do.²⁷ Of course, Hayek disagreed with Keynes over this. But, this is essentially, for us, a dispute within this tradition of liberalism. It is a dispute over what it means to get the rules of the game ‘right’. Nevertheless, we make no strong claim in this regard here because our paper has a different objective: an assessment of his institutional model of growth.

His argument in The Good Society depends in important respects on this institutionalist model of growth. We claim this model is both plausible and distinctive (see the second section, notably on the need for property rights to evolve if growth is to be sustained because the appetite for risk taking is endogenous). He also used his analysis to make some predictions regarding the kind of institutional change that would be necessary for the revival of liberalism. These predictions are, in a broad brush sense, consistent with much of the institutional change that accompanied the revival of liberalism in the post-war period (see the third section). In short, he uses his institutional model to do two things: (a) generate a novel and plausible explanation of the rise in security and stagnating living standards that were associated with the challenge to liberalism in the 1930s and (b) to make some predictions for how to restore growth and reduce insecurity (and hence provide for the revival of liberalism) that are broadly consistent with what happened with the revival of liberalism in the post-war period. These are two reasons for taking Lippmann's institutional analysis seriously.

We conclude by developing our third argument for commending Lippmann: his model of growth might provide insights with respect to the contemporary stagnation in living standards and increased insecurity that many rich countries have experienced. Lippmann's model of growth alerts policy makers to the sunset quality of any set of property rights. If growth slows and insecurity rises, then it is likely for Lippmann to be due to the failure of property rights to adjust either in their encouragement to risk taking and/or in their check on sources of unequal bargaining power. The novelty of this argument in relation to the current discussion of stagnating living standards is in the first aspect.²⁸ We focus, therefore, briefly on how property rights might be adjusted to encourage risk taking.

In principle, there are two policy strategies. One makes risk taking more attractive by increasing the expected return: e.g. by reducing the tax on upside gains. The other reduces the perceived risks: e.g. through innovative forms of insurance. Although it is difficult to make simple cross country comparisons, the influence of both strategies would seem to be apparent in the contrasting experiences of rich countries during the post war period. For example, consider the experiences of, say, the US and the Nordic European countries. There is a plausibly greater relative emphasis on expected returns in the US, as income inequality is greater and welfare state is smaller there as compared with the Nordic economies. In contrast, there is greater relative emphasis on reducing risks through more compressed income inequality and a larger welfare state in Nordic countries. Unless people responded to both expected returns and risk, it would be difficult to understand in this simple comparison why the US and Nordic countries had similar records of productivity growth over the whole period.

In general, therefore, it is an empirical question as to which strategy might prove the most effective to pursue now. There are two reasons, however, for supposing that contemporary policy should be focussed on reducing risk. One is that the approach of improving the expected returns has been tried and found wanting. Top tax rates have been reduced in the OECD, inequality has increased and it seems not to have stemmed the slowdown in productivity growth. Lippmann would not have been surprised, not least because the other source of slow productivity growth, unequal bargaining, is liable to be entrenched by such policies. Second, the OECD (2015) and IMF (2017) have both argued, in part on empirical grounds, that a reduction in inequality will likely boost productivity growth in these rich countries now. Thus, innovative forms of insurance to reduce risk look the more promising contemporary policy approach.

Lippmann's analysis contains an important insight regarding such innovations. Recall that, on his account, the perception of risk tends to increase with productivity growth because the latter generates complexity which, in turn, is a source of unpredictability – i.e. background risks in the sense that they cannot be readily tied to anyone's actions. These are not the kinds of risks that are well suited to private insurance (Barr 2001; Feduzi and Runde, 2011; Sinn, 1995). It is in the nature of the genuinely unpredictable aspects of what happens in decision making that contracts cannot be written ex ante to cover such contingencies. Instead, the best that can be done is to provide forms of social insurance that limit the possible downsides by providing ex post compensation for losses that will sometimes arise and which could not have been anticipated ex ante. A state job guarantee, a basic income and a flat tax with a higher tax free allowance are examples of current changes in the positive aspects of property rights that could reduce downside risks.²⁹

In short, Lippmann's institutional analysis is distinctive in several respects, and if he is right, it contains important policy advice for the present: increasingly, the rights regime will have to develop to encourage risk taking through the provision of so-called ‘positive’ rights through new forms of social insurance.³⁰