I. Introductionweb.stanford.edu/~saumitra/papers/JhaQJE2015.pdfSaumitra Jha The English Parliament’s struggle for supremacy against monarchical dic-tatorship during the Civil War

FINANCIAL ASSET HOLDINGS ANDPOLITICAL ATTITUDES: EVIDENCE FROM

REVOLUTIONARY ENGLAND*

Saumitra Jha

The English Parliament’s struggle for supremacy against monarchical dic-tatorship during the Civil War (1642–1648) was crucial for the establishment ofrepresentative government, yet its lessons continue to be debated. I exploitnovel data on individual MPs drawn from 1,842 biographies to show that theconflict was over overseas interests and other factors over which the executiveenjoyed broad constitutional discretion, rather than over domestic propertyrights. I further exploit the coincidence of individual MPs’ ability to sign legallybinding share contracts with novel share offerings by overseas companies tomeasure the effect of overseas share investment on their political attitudes.I show that overseas shareholding pushed moderates lacking prior mercantileinterests to support reform. I interpret the effect of financial assetholding asallowing new investors to exploit emerging economic opportunities overseas,aligning their interests with traders. By consolidating a broad parliamentarymajority that favored reform, the introduction of financial assets also broadenedsupport for the institutionalization of parliamentary supremacy over dictatorialrule. JEL Codes: O10, G11, F10, K00, N23, P10.

I. Introduction

The seizure of executive authority by Parliament from theCrown in the years spanning England’s Civil War (1642–1648)has been called the ‘‘fountainhead of revolutions’’ (Schama 2001).Before the Long Parliament (1640–1660), England approximateda dictatorship. The Crown called and dismissed Parliaments at

*An earlier version circulated as ‘‘Financial Innovations and PoliticalDevelopment.’’ I owe particular thanks to Ran Abramitzky, Susan Athey, AvnerGreif, Aprajit Mahajan, Elhanan Helpman, three referees and Amrita Ahuja, KenArrow, Dan Bogart, Ernesto Dal Bo, Ann Carlos, Mauricio Drelichman, JeffFrieden, Sean Gailmard, Oscar Gelderblom, Claudia Goldin, Luigi Guiso, EricHilt, Kimuli Kasara, Keith Krehbiel, Dorothy Kronick, Peter Koudijs, TimurKuran, David Laitin, Jessica Leino, Kris Mitchener, Noel Maurer, Ted Miguel,Pedro Miranda, Larry Neal, Francisco Perez-Gonzalez, James Robinson, TomRomer, Matthias Schundeln, Carmit Segal, Jordan Siegel, David Stasavage,Nathan Sussman, Barry Weingast, and audiences at Berkeley, Brown, GMU,Harvard, LSE, MIT, NYU, Penn, Princeton, Stanford, the World Bank, Yale,and numerous conferences. Zac Peskowitz and Peter Schram provided excellentresearch assistance. This research benefited from articles made available prior topublication by the History of Parliament Trust, and financial support fromSIEPR, the Harvard Academy, CSDP, and the Niehaus Center at Princeton.

! The Author(s) 2015. Published by Oxford University Press, on behalf of Presidentand Fellows of Harvard College. All rights reserved. For Permissions, please email:[email protected] Quarterly Journal of Economics (2015), 1485–1545. doi:10.1093/qje/qjv019.Advance Access publication on May 6, 2015.

1485

Dow

nloaded from https://academ

ic.oup.com/qje/article-abstract/130/3/1485/1934276 by Stanford M

edical Center user on 20 Septem

ber 2019

will. Between 1629 and 1640, no Parliament met at all (see FigureA1.1 in the Online Appendix).

The summoning of the Long Parliament in 1640 precipitateda process of revolutionary change that inspired future revolutionsin both the Old World and the New (Carlyle 1897 [1845]; Stone1985; Pincus 2009). Reformers were able to consolidate a majorityin the House of Commons and broaden support outside it. Thisallowed Parliament, through both legislative and ultimately vio-lent means, to acquire new rights to convene without royal ap-proval, control state finance, and direct foreign policy and war.These constitutional changes set England on a path of experimen-tation that led to one of the world’s most enduring institutions ofrepresentative government.

The central puzzle—what drove the creation of a novel,broad, and ultimately successful coalition in favor of parliamen-tary supremacy?—has been a focus of much debate ever since.Economists have focused on three key explanations. North andWeingast’s (1989) groundbreaking work argues that a coalitionformed to defend property rights in response to a political shockin the form of excessive executive greed by the Stuart monarchs.The successful removal of kings yielded a credible threat thatenabled future rulers to commit not to expropriate property, lead-ing to dramatic financial and fiscal development in England.Others suggest that economic shocks created new commercialmiddle classes that then sought to protect their newly acquiredwealth from executive predation. For Karl Marx, England’s CivilWar was the ‘‘First Bourgeois Revolution’’ (Stone 1985).Acemoglu, Johnson, and Robinson (2005) suggest that thiswealth was acquired by new merchants involved in trade acrossthe Atlantic (see also Brenner 1993). Rajan and Zingales (2003),Moore (1993 [1966]), and Tawney (1941) propose, in contrast,that the revolution was led by newly commercialized gentrythat acquired land due to the dissolution of monasteries in1536–1541.

In all three interpretations, a constituency of wealth holdersemerged that demanded and obtained improved protection of do-mestic property rights, leading both to representative govern-ment and to economic growth. However, all these theories haveproven difficult thus far to reconcile with a body of indirect em-pirical evidence that suggests that domestic property rights werealready relatively secure in seventeenth-century England, atleast for the wealthy, and did not experience much change

QUARTERLY JOURNAL OF ECONOMICS1486

Dow




ber 2019

http://qje.oxfordjournals.org/lookup/suppl/doi:10.1093/qje/qjv019/-/DC1

thereafter (e.g., Clark 1996; Quinn 2001; Sussman and Yafeh2004; de Lara, Greif, and Jha 2008).

In this article, I provide the first direct evidence on the im-portance of different endowments on individuals’ decisions tochoose to support the coalition in favor of monarchical dictator-ship or parliamentary supremacy during the English Civil War.Using novel data on economic and social endowments drawn from1,842 biographies of political elites, including the paternal line-ages of 528 members of the Long Parliament that initiatedEngland’s revolutionary reforms, I first address the central ques-tion: who rebelled? I show, consistent with a simple theoreticalframework, that those endowments most subject to executive dis-cretion under the existing constitution were most likely to influ-ence an individual’s decision to rebel. In contrast, a range ofendowed measures of domestic wealth, that themselves stronglypredict contemporary income, have no effect on the propensity ofan individual to support parliamentary supremacy. These mea-sures include whether an individual was an heir, stood to inheritlanded estates, or had a father with a peerage, baronetcy, orknighthood. These null effects for endowments that were majorcomponents of elites’ portfolios suggest that domestic wealth,whether old or newly acquired, was not expected to experiencelarge decreases in expropriation risk. Thus it is unlikely that acoalition of domestic wealthholders chose to support revolution-ary change during the Civil War, as suggested in all three ofthe most prominent economic interpretations of the Englishrevolution.

How then was the broad coalition in favor of parliamentarysupremacy formed across groups with initially divergent inter-ests? I provide evidence that a new type of financial asset—sharesin overseas joint stock companies—played an economically andstatistically significant role in transforming fragmented interestsinto a broad coalition that favored parliamentary supremacy.Joint stock companies had become popular in the late sixteenthcentury to share risks and opportunities overseas. I argue thatsince the rights needed to profit from overseas investment be-longed to and were benefiting the Crown at investors’ expense,the introduction of shares aligned the incentives of a broad coa-lition in favor of constitutional reforms aimed at seizing control ofthese rights. In particular, shares aligned the interests ofnonmerchants, who otherwise would have lacked exposure to lu-crative opportunities and expropriation risks overseas, with

FINANCIAL ASSET HOLDINGS AND POLITICAL ATTITUDES 1487

Dow




ber 2019

merchants, who could trade on their own behalf in the absence ofshares. I show that shares shifted the views of moderate nonmer-chants and consolidated a majority of core support for reformersin Parliament. Thus, the broadening of the coalition was condu-cive not merely for reforms beneficial to narrow interests but alsofor parliamentary supremacy.

Empirically identifying the effects of financial assetholding isnaturally very difficult given the selection processes throughwhich investors choose to hold and choose to retain financialassets. To the best of my knowledge, this is the first article toexploit plausibly exogenous variation to estimate the effect of fi-nancial shareholding on the political attitudes of individualelites.1

I use two complementary means to estimate the causal effectof overseas shareholding. First I match investor MPs to nonin-vestor MPs along a range of endowed wealth and locational char-acteristics likely to explain deviations from the mean-varianceefficient benchmark—where all agents hold the market portfo-lio—to construct lower bound estimates of the effect of sharehold-ing. I then identify the effect of shareholding using a series ofplausibly exogenous shocks to the propensity to invest overseasamong individuals who turned 21 and were able to sign legallybinding share contracts just as overseas companies were makingan initial public offering (IPO) of shares. I detail each approach inturn.

First I show that among members of Parliament, an elitegroup of frequent visitors to the financial capital, those whoheld shares appear actually very similar to others across arange of endowments that might explain differences in propensi-ties to hold shares in broader populations (Tables I and II). Notsurprisingly, then the measured effect of shareholding on sup-port for parliamentary supremacy of 21.5 percentage points isrobust to matching on a range of relevant individual wealthand geographical endowments and constituency characteristics,reflected in a preferred lower bound estimate of around 20 per-centage points (Table III). I further test for a differential

1. I build on an emerging literature in finance, with arguably the closest paperbeing Kaustia, Knupfer, and Torstila (2015). They exploit a regional matched com-parison of Finnish areas where mutual funds became publicly listed companies toothers where they did not and document a relationship between the extent ofdemutualization in an area and the proportion of right-of-center votes.


Dow




ber 2019

alignment effect of shares in allowing nonmerchants to share inthe exposure to new overseas opportunities already enjoyed bymerchants. I find that while MPs with mercantile endowmentswere 18 percentage points more likely to rebel regardless ofshares, the effect of shares specifically shifts the political alle-giance of nonmerchants (Table IV and Figure II).

I evaluate alternative explanations and mechanisms. I showthat overseas investors were not any richer at the time of the CivilWar, suggesting that a shock to existing wealth was not drivingtheir support for reform. In fact, investors in overseas companiesthat were unprofitable prior to the Civil War, and thus had less tolose from regime change, were more likely to rebel. Unobserveddifferences in risk preferences alone also do not explain the re-sults: investors in domestic joint stock companies that were alsorisky—but faced less executive risk—did not rebel. Similarly, itdoes not appear to be the case that overseas investors happenedto have a preexisting antimonarchical vision. Instead overseasinvestors were actually more likely to cultivate ties to the courtprior to the Civil War, not less, consistent with a strategy of se-curing overseas investments through political patronage.

As in all quasi-experimental settings, the possibility that re-sidual unobserved selection biases these results remains. I there-fore exploit a series of plausibly exogenous shocks to thepropensity to invest overseas among agents who attained adult-hood just as companies were making an IPO of shares. Turning21 allowed an MP to buy and sell property as well as sign legallybinding contracts, including share contracts, which he could nothave done had he been 20. Gaining this freedom to contract andreallocate portfolios during moments of broad enthusiasm foroverseas ventures that accompanied the creation of new compa-nies also increased an MP’s propensity to invest even relative tothose who were 22 or older at the time of an IPO, and thus werenot legally constrained from investing (Figure IV). I show thatMPs are around 22.3 percentage points more likely to invest inoverseas joint stock companies if they came of age during an IPOyear than those that came of age at other times within a five-yearwindow of an IPO, and around 15.9 percentage points more likelyto support parliamentary supremacy in the Civil War. This trans-lates into an increased probability of rebellion of around 71.2 per-centage points among those who invested in shares because theycame of age in an IPO year (Table VIII). Furthermore, consistentwith an alignment effect of shares, while both merchants and


Dow




ber 2019

nonmerchants are more likely to invest if they came of age in IPOyear, the effect of holding overseas shares once again onlychanges the political allegiance of nonmerchants (Fig. III andTable VIII).

In fact, shares appear to have switched moderate nonmer-chants from supporting the Crown to favoring rebellion, consoli-dating a majority in Parliament, and arguably influencedmoderate elites outside Parliament as well (Figure VI). The com-pliant switchers—investor MPs who supported parliamentary su-premacy but were predicted to be royalists in the absence ofshares—may have encompassed four of the six members thatdrafted the Grand Remonstrance (the manifesto forParliament’s reforms on the eve of war) and all of the FiveMembers whom the king identified as ringleaders of the reforms(Online Appendix Table A1.13). Far from being extremists, thesemembers appear to have been actually more likely to supportmonarchical dictatorship had they not invested in overseasjoint stock companies.

Beyond the stark choice to support Crown or parliamentarycontrol of government during the violence of the Civil War, I tracea direct link between overseas share ownership and support forconstitutional reform throughout the life cycle of the early strug-gle for parliamentary supremacy. I exploit surviving records thatindicate support for reform during early legislative attempts tochange the constitution, on the eve of the outbreak of violenthostilities, and in the parliament of post–Civil War victors thatimplemented dramatic investments in England’s navy in defenseof overseas trade. A consistent picture emerges: the introductionof overseas shares appears to lead nonmerchants to make similarpolitical decisions to those with prior overseas trade interests,consolidating a majority coalition that favored revolutionaryreform.

My findings support, though add further nuance to theAcemoglu, Johnson, and Robinson (2005) interpretation ofEngland’s transition: new Atlantic trade routes did matter.However, England did not become an oligarchy of newly enrichedtraders; in fact the Crown extracted many of the gains from newtrades (Online Appendix Figure A1.2). Instead, the introductionof shares in these new opportunities promised future gains to abroader coalition, transcending conflicting initial interests suchas those between merchant and nonmerchant, devout andworldly. The fact that investor-reformers were able to consolidate


Dow




ber 2019



a broad majority within Parliament and enjoyed the support ofmoderates outside of it (Figure VI) explains why reformers inEngland pushed for fundamental institutional change aimed atparliamentary supremacy rather than engaging in narrower sec-tarian or class conflict.

By identifying a new mechanism to explain a pivotal momentin England’s development, this article relates to important liter-atures in finance and the political economy of development. Muchblame for underdevelopment around the world has been attrib-uted to a failure to align the incentives of disparate interestgroups in favor of political reform and beneficial public policies(e.g., Haber and Perotti 2008; Rajan 2009; Benmelech andMoskowitz 2010; Acemoglu and Robinson 2012). Unlike anumber of papers in historical political economy, where interestsform around differences in wealth, and shocks to existing wealthprovide the impetus for change, I provide evidence that shocks tofuture opportunities for wealth, in combination with the intro-duction of financial assets that allowed initially disparategroups to share in those opportunities, played an important rolein building coalitions for reform. The introduction of financialassets to share in future opportunities for wealth is arguablymuch easier to influence through policy than overcoming theloss aversion induced by attempts to redistribute existingwealth to create common middle class interests (Jha 2012).Furthermore, in England, important positive feedback existedbetween financial development and political institutionalchange. As I show, the introduction of financial assets helpedbroaden access to political power through the alignment of inter-ests favoring parliamentary supremacy. The fact that parliamen-tary supremacy led in turn to the subsequent broadening ofaccess to economic opportunities through England’s financial rev-olution (North and Weingast 1989) suggests an intriguing role forfinancial assets in fostering the coincidence of inclusive economicand political institutions that have been credited for economicdevelopment more generally (Greif 2005; Acemoglu andRobinson 2012).

I begin by providing relevant context for my interpretation,supported by novel statistical evidence in the Online Appendix. Asimple framework tracing the relationship between MPs’ endow-ments and choices motivates the empirical methodology. I thenintroduce a new micro data set drawn from 1,842 parental andpolitician biographies and present the results. I conclude by


Dow




ber 2019


discussing the aftermath of the struggle and the broaderimplications.

II. Context

Inspired by Spanish and Portuguese successes in exploitingnew sea routes to the New World and Asia, the first English jointstock company was founded in 1552 (Online Appendix Table A1.1provides a timeline of key financial and political events). The ex-plicit purpose of issuing shares in what became the RussiaCompany was to enable merchants to share the risks of findinga sea route to the Indies with nonmerchants (Hakluyt 1962[1589], p. 267).2 For the first time, joint stock companies enablednonmerchants to take advantage of potentially highly lucrativeoverseas opportunities, without themselves specializing in navi-gation and commerce. Dividends were promised both in terms offuture profits and in terms of land and property rights in the newterritories.3

English overseas ventures languished, however, untilSeptember 1580, when the Pelican, the sole survivor of a fleetof five ships that had sailed three years earlier, moored inPlymouth. Her captain, Francis Drake, had achieved an unlikelysuccess—the circumnavigation of the world, direct trade with theSpice Islands, and a raid on Spanish treasure ships in the Pacific.Drake’s voyage and the charts of watering places, ports, andtrade routes he constructed meant that for the first timeEnglish traders had the vital intelligence necessary to breakinto Portuguese and Spanish monopolies in Eastern trades

2. The joint stock company was innovative in a number of ways. First, largenumbers of individuals, particularly nonmerchants, could invest in shares andoverseas opportunities for the first time. Second, unlike traditional overseas regu-latory companies where merchants gained the freedom after long apprenticeshipsto tradeon their own account or in small partnerships, now trade was doneon behalfof the company. Third, England’s early joint stock companies were run by courts ofdirectors who were elected by votes allocated in proportion to the votes held. Theselatter two factors meant that joint stock companies had a system of governancedesigned to accommodate larger groups of investors.

3. For example, each £12.50 share in the Virginia Company also yielded itsowner 100 acres. Even companies, such as the Bermuda Company, which had al-ready distributed their land, or the Gynney Bynney Company, which did not colo-nize and wrapped up operations, retained certain property rights, such as amonopoly over the right to trade with the colony or region, that had also promiseddividend streams (Scott 1910, p. 185).


Dow




ber 2019


(Andrews 1967). Not only did this voyage change the feasibility ofdirect trade, it also amply demonstrated the scale of profits to behad.4 In 1585, Drake successfully raided Spanish ports in theAtlantic as well, demonstrating their vulnerability to English at-tacks. His extraordinary achievements made him a national heroand triggered great enthusiasm for overseas joint stock ven-tures.5 Indeed, mentions of trade, Indies, and the Americas inpublished tracts in English experienced particularly largeboosts in 1580 and 1585, thereafter achieving levels comparableto and often exceeding written mentions of Pope, Catholic, papist,and bishop and of rights, privileges, liberties, and freedoms in theyears preceding the Civil War (Online Appendix Figure A1.4).Joint stock companies were established to trade with and colonizemuch of the world, including Africa, the Levant, the East Indies,Virginia, New England, and Latin America (Online AppendixTable A1.2). Enthusiasm for shares in these companies spreadbeyond merchants to encompass a broad spectrum of politicalelites. The more than 6,366 investors between 1575 and 1630included 23 percent of all members of Parliament seated in thatperiod (Rabb 1967) (see also Online Appendix Figure A1.5).

The incentives generated from overseas share investmentsalso tended to be stronger than those faced by modern-dayholders of common stock. Without limited liability, joint stockinvestors faced long-term obligations should the company needfurther funds.6 Secondary markets for shares also only emergedin the 1660s, after the Civil War (Carlos, Key, and Dupree 1998;Harris 2009). With divestment difficult, individuals were legallyand financially locked into the fate of the companies in which theyhad invested.

Despite the enthusiasm for investment, most joint stock ven-tures failed to make profits prior to the Civil War. An important

4. The Spanish ambassador, Bernardino de Mendoza, estimated the profitsfrom Drake’s voyage to be worth 1.5 million pesos. Elizabeth I alone receivedaround £64,600 of gold and silver (Kelsey 1998, p. 215).

5. Mendoza cautioned Philip II to destroy all English and French ships enter-ing the Pacific as ‘‘at present there is hardly an Englishman who is not talking ofundertaking the voyage, so encouraged are they by Drake’s return . . . everybodywants a share in the [next] expedition’’ (Rabb 1967, p. 20).

6. For example, between 1554 to 1586, the Russia Company paid no dividendsto its shareholders but continued to make calls on them for funds to pay for its highcosts (Harris 2000, p. 44). General limited liability did not emerge until 1856(Harris 2000, pp. 127–128).


Dow




ber 2019





reason was that unlike domestic property rights that were gov-erned by common law precedent, foreign trade was governed bycivil law, administered by the Crown in the Admiralty courts(Gardiner 1862, p. 87; Burgess 1992). English rulers also couldgrant and revoke charters to companies overseas, impose cus-toms, and create monopolies of newly introduced goods, as com-merce and innovation was believed to be protected by the king’sforeign policy (Gras 1912). As Online Appendix Table A1.2 docu-ments, prior to the Civil War, even those joint stock companiesthat enjoyed initial profits faced not only foreign predation butalso Crown expropriation through rising customs charges or therevocation of their charters.7 These setbacks eventually led to adecline in enthusiasm for investment in the 1630s (Rabb 1967, pp.71–75). In contrast, the contribution of overseas customs to totalCrown revenues rose from 5.2 percent in 1552 to 52.5 percent onthe eve of the Civil War in 1642 (Online Appendix Figure A1.2).The increase in value of the Crown’s sovereignty rights over cus-toms and foreign policy meant that England’s kings, though stillrelatively limited in their ability to extract resources domesti-cally, were becoming increasingly enriched over time.8 At thesame time, even shareholders of companies whose charters hadbeen revoked, like the Virginia Company, continued to possesslegal obligations and claims on hundreds of acres of undevelopedterritory overseas with the potential to rapidly gain value in thefuture, a potential that was indeed met after the Civil War(Online Appendix Figure A1.11).

Not surprisingly then, attempts to bargain over the controlover rights over customs and foreign policy played a pivotal rolein parliamentary debates from 1603 to 1625, with joint stock in-vestors playing prominent roles (Hill 1961, pp. 48–50; Rabb1998). These culminated in the Great Contract of 1610, an

7. Even the fortunes of the East India Company were reversed in 1623, whenthe Dutch destroyed their factory at Amboina in the Spice Islands, with the Crownalso raising customs on pepper and extracting loans. Outbound shipping declinedand the company discussed terminating itself on the eve of the Civil War(Chaudhuri 1965).

8. That the monarch was the major beneficiary of expanded overseas trade,and probate records of individuals’ estates show no evidence that the wealth ofEngland’s business community rose in any part of the distribution until the tradeboom after the Civil War (Grassby 1970), suggests that the enrichment of Atlantictraders proposed by Acemoglu, Johnson, and Robinson (2005), while potentiallyshaping political attitudes prior to the Glorious Revolution, did not play that role inthe Civil War.


Dow




ber 2019




attempt to exchange the king’s authority over independentmeans of revenue for assured parliamentary grants of directtaxes. Bargaining foundered, however. Parliament wanted con-trol over customs, and the king was counseled not to surrender‘‘the fairest flowers for profit and command in all his garland’’(Smith 1999, pp. 53–55). Customs revenue helped the Crownrely less on Parliament, with James I dismissing the AddledParliament of 1614 and ruling without parliamentary taxes fora 11-year period (see Online Appendix Figures A1.1 and A1.2).

Reconciliation could and did occur. The Happy Parliament of1624 witnessed a spike in acts of Parliament passed (OnlineAppendix Fig. A1.1), with Charles I willing to accept parliamen-tary grants of taxation explicitly tied to the declaration of warwith Spain (Hill 1961, p. 58).9 However, disputes over foreignpolicy soon reemerged. Rather than fighting a ‘‘profitable war’’against Spanish colonies in the West Indies as investor-reformerspreferred (Mitchell 1957), the king chose to invade the Spanishmainland. With Parliament refusing to grant taxes to fund such astrategy, the king once again dismissed Parliament in 1629 andgoverned England as a dictator for the next 11 years.

Though the king’s resources were growing (Online AppendixFigure A1.2), invasion by the Scots overwhelmed his finances andled Charles to summon the Long Parliament in 1641. Over thenext two years, the Long Parliament passed, and the king signed,key constitutional reforms. These included the abolition of theroyal prerogative courts and the passage of the Triennial Actthat guaranteed that Parliament must be called at least everythree years and, critically, could not be dismissed without itsown consent. Parliament also passed bills to arrest the king’schief councillors, including the Earl of Strafford, and sought con-trol over future appointments.10

9. The collegiality of 1624 is often cited by revisionist historians as evidenceagainst broad socioeconomic explanations for England’s revolution (e.g., Morrill1993). However, it is also consistent with my interpretation: reformers’ willingnessto support the monarch rose when he was willing to cede prerogative control overforeign policy to Parliament (see also Smith 1999.)

10. Unlike later in the seventeenth century, records of MPs’ votes in Parliamentwere not systematically kept, with one exception. InApril 1641,Parliament voted toconvict the king’s chief advisor, Thomas Wentworth, the Earl of Strafford. Thosewho voted against the conviction were seen as supporters of royal authority andtheir names were anonymously posted in Westminster Yard, providing a usefulprewar indicator of support.


Dow




ber 2019







In November 1641, investor-reformers penned the GrandRemonstrance, a manifesto aimed at instituting parliamentaryauthority over remaining Crown rights, including over foreignpolicy, finance, and the armed forces. This led the king to ille-gally enter the House of Commons to arrest the Five Membersconsidered the ringleaders of the parliamentary opposition inJanuary 1642. Parliament summoned London’s citizen militiain its defense. The king abandoned the city and, in June 1642,raised his war banner, threatening to use force in defense ofCrown rights.11

Sufficient support in 1628 had enabled reformers to deny theking grants of taxation. In the Long Parliament, reformers alsocommanded enough support to push through key legislative re-forms. However, it was less clear whether this support was strongenough and broad enough to persist when individuals were forcedto choose and to commit themselves publicly and likely irrevers-ibly to the cause of revolutionary change or to accede to reinvigo-rated dictatorial rule. It is to understand the determinants of thatdecision that I now turn.

III. Empirical Strategy

Since the English revolution was ultimately a battle betweenthe supremacy of the king and Parliament, theories of whopushed for parliamentary supremacy should find validationamong those who sat in the Long Parliament and initiated theconflict. Despite the importance of this episode, this is, to my bestknowledge, the first article to systematically gather, measure,and analyze the endowments and political choices of the individ-ual MPs who initiated England’s constitutional reforms.A particularly valuable feature of these data is that the political

11. Eikon Basilike, purportedly written by Charles I himself, explains his ulti-mate rationale for war: ‘‘For although I can be content to eclipse my own beams tosatisfy [Parliament’s] fears . . . yet I will never consent to put out the sun of sover-eignty to all posterity and succeeding kings, whose just recovery of their rights fromunjust usurpations and extortions shall never be prejudiced or obstructed by anyact of mine’’ (Charles I and Gauden 1649, pp. 48–49). If there were discontinuousdifferences in the ceding of prerogative rights in shaping future bargaining power,Charles’s rationale would mimic conditions for the failure of the political Coasetheorem (Fearon 1996).


Dow




ber 2019

allegiances of virtually all members were publicly revealed bytheir actions during the Civil War. Parliamentary fence-sitterson the eve of the Civil War were forced to choose between accept-ing the conflicting summons of the king to Oxford and ofParliament to Westminster. By 1644, every living parliamentar-ian could be associated with one side or another (Brunton andPennington 1954) (Online Appendix 2).12

A simple theoretical framework can shed light on the rela-tionship between endowments, property rights, and politicalchoices. Suppose that the utility for a member of parliamentcan be summarized by the following additive relationship:

Ui ¼X

j

xij�jjz þ uz;ð1Þ

where xij are predetermined individual endowments,�jjz; z 2 fP;Rg denotes the rate of return on endowment j inthe state of the world where either the monarchy (R) orParliament (P) wins the civil war, and uz; z 2 fP;Rg containsother orthogonal factors that influence expected utility in thestate of the world z. Suppose that each agent believes that withprobability �, Parliament (P) will win the struggle against royalauthority (R).13 Then the expected utility for each MP is:

EUi ¼ �X

j

xij�jjP þ uP

!þ ð1� �Þ

Xj

xij�jjR þ uR

!:ð2Þ

Suppose further that the support of an individual agent in-creases the chances of victory by an amount s>0. Then the

12. The well-documented allegiances of parliamentarians differs from otherpublic figures. The loyalties of local leaders were confounded by both local powerpolitics and the presence of occupying armies. Even London, often seen as the epi-center of parliamentary power, initially had a strong royalist presence among itsleadership, including a royalist Lord Mayor.

13. Naturally, � is likely to be affected by other agents’ choices. For plausiblespecifications of the multiagent game, multiple values of � will be consistent withequilibrium. However, as will be shown, the specific realization of� is irrelevant foran agent’s decision, as long as �± s is interior. This condition—that there is someuncertainty about whether Parliament or monarchy wins regardless of an individ-ual agent’s choices—makes sense in the historical context.


Dow




ber 2019


agent’s problem is to choose to support parliamentary supremacyor royal control:

maxz2fP;Rg

"ð�þ sÞ

Xj

xij�jjP þ uP

!þ ð1� �� sÞ

Xj

xij�jjR þ uR;

!;

ð�� sÞX

j

xij�jjP þ uP

!þ ð1� �þ sÞ

Xj

xij�jjR þ uR;

!#:

The optimal choice naturally implies a cut-off strategy: an agentwill support parliamentary supremacy if:

sX

j

xij½�jjP � �jjR� þ ðuP � uRÞ

!> 0:ð3Þ

Inequality (3) yields an implicit condition on the minimumuP � uR required for support for parliamentary control. Thus theprobability of supporting Parliament is:

PfPg ¼ FX

j

sxij½�jjP � �jjR�

!;ð4Þ

where Fð�Þ is the cumulative density function of uP � uR.Notice that if MPs believe that their support has a nonzero

effect on the outcome (s>0), the influence of each endowment isdetermined by xij½�jjP � �jjR�.

14 In particular, even small differ-ences in ½�jjP � �jjR� will be magnified for those endowments xij

that are large components of an MP’s portfolio, such as landedestates. For a large endowment to have no effect, it must be there-fore that �jjP � �jjR: that is, that the expected return on the en-dowment is unaffected by regime change. This is consistent withthe presence of secure property rights under the existing consti-tution. As I shall show, a series of endowments that are likely tobe major drivers of wealth and strongly predict contemporaneousincome have no effect on the propensity to rebel.

Similarly, equation (4) implies that even if an asset endow-ment xij was small in current value, such as was likely the case for

14. Observe that if s = 0, all endowments should be irrelevant, and thus a jointtest of the significance of all endowments should be zero. The regressions that followare sufficiently significant to reject this test.


Dow




ber 2019

overseas share investments in unprofitable companies, the ex-pected change in return ½�jjP � �jjR� must be large for that assetto be influential in determining an MP’s propensity to supportregime change. Such would be true for overseas property rightsand other endowments heavily exposed to executive discretionunder the existing constitution.

I estimate equation (4) using:

Rebeli ¼X

j

gj ~xij þ �Merchanti þ �OverseasInvi þ ei;ð5Þ

where Rebeli is an indicator for whether an individual sup-ported Parliament in 1642–1648, and all endowments are mea-sured prior to those dates.15 I also separate the matrix ofendowments Xj into Merchanti, an indicator for whether an in-dividual was apprenticed to a merchant company when youngor whose father was a merchant, and OverseasInvi, an indicatorfor whether the individual invested in overseas joint stock priorto 1640.

The coefficients fgj; �; �g measure how a unit change in eachendowment changes the probability of support for parliamentarysupremacy over dictatorial rule. From equation (4), they can alsobe interpreted structurally as sð�jjP � �jjRÞ, the differential ex-pected rate of return on each endowment from parliamentarysupremacy over dictatorial rule scaled by the influence an MPbelieves his support will have on the chances of victory.16

To obtain a causal estimate of the effect of overseas share-holding in particular, I use two complementary approaches. Thefirst approach compares shareholder MPs to noninvestor MPsmatched along a range of those endowed wealth and locationalcharacteristics that have been emphasized to explain the stock-holding puzzle—departures from the mean-variance efficientportfolio where every agent should hold shares—and conditionalon which the decision to hold shares could plausibly be idiosyn-cratic and as good as random (e.g., Merton 1987; Coval and

15. The Civil War itself was unexpected, even as late as 1641 (Whitelocke 1732[1682], p. 60).

16. Note that those more able to sway the ultimate outcome (s ") also experiencea greater magnification of small differences in ½�jjP � �jjR�. Thus, since MPs’ influ-ence on the outcome likely exceeded that of non-MPs, the MP estimates imply anupper bound estimate for non-MPs within the framework. In particular, a lack of aneffect of domestic wealth endowments for MPs also suggests that domestic wealthshould not influence support by non-MPs.


Dow




ber 2019

Moskowitz 1999; Guiso and Jappelli 2005; Barber and Odean2006).17

In particular, the literature on the stockholding puzzle em-phasizes two sets of transaction costs that lead to deviations fromthe efficient portfolio. First, it is plausible, particularly in a na-scent stock market environment, that even fully rational agentscannot invest in particular assets that they do not know exist(Merton 1987). If investors face fixed costs in gaining access toinformation about particular assets, or more generally in pur-chasing them, there will be a correlation between the possessionof liquid assets, or wealth, and investment in stocks. I thereforeinclude in Xj a set of controls for wealth endowments. As dis-cussed already, these include whether an agent was the eldestson (and thus the heir, particularly important during this periodduring which primogeniture was customary), whether the agentinherited land or manors, whether the MP experienced wardship,and whether the agent’s father had an aristocratic title, a baron-etcy, or a knighthood. Fixed costs in learning about shares mayalso lead to local geographical concentrations among share-holders (Coval and Moskowitz 1999; Zhu 2002). Thus I includein Xj a set of controls for the proximity to London, already themajor financial center, of the residences of the father of each MP,as well as the distance to the constituency of the MP, and compareMPs who represent constituencies within the same county.

Though this theory-driven choice of wealth and geographicalcontrols cannot cover all the potential avenues by which

17. In the canonical model of portfolio choice, assuming that investors areaware of all assets, there are no transaction costs, and there are no uninsurablerisks (such as accumulated human capital), all agents seeking the mean-varianceefficient portfolio should choose to hold the market portfolio. The market portfoliowould include shares in joint stock companies, as soon as they are introduced.Differences in risk preference affect the allocation of assets between risky andnonrisky assets, but not the particular set of risky assets, such as shares in jointstock companies (Guiso and Jappelli 2005). Thus in a frictionless environment,there should be no systematic selection bias among those who choose to invest inshares, and OLS will provide unbiased estimates of the average effect of the intro-duction of shares on support for parliamentary reform among investors.Furthermore, if investors’ utility functions obey constant absolute risk aversion,investors will invest the same amount in stock. Even with constant relative riskaversion, asset shares will be independent of wealth: rich and poor will hold thesame proportion of each asset, including stock, but in differing amounts (Guiso andJappelli 2005).


Dow




ber 2019

shareholders may have been selected, they do allow me to matchinvestors to noninvestors based on those transaction costs consid-ered most important. Assuming that any residual unobservabledimensions that might influence an individual’s propensity tohold shares have similar magnitude effects as the set of theory-driven observable controls, I construct benchmark lower boundestimates of the effects of shareholding. I then present a comple-mentary strategy that relaxes the assumption that unobservableshave effects that are similar in magnitude to observables, andinstead use the plausibly exogenous coincidence of an individual’sability to sign share contracts at age 21 with the IPO of overseasshares in a company in that year to identify the effects of overseasshareholding on support for parliamentary supremacy.

Differences in human capital endowments can present animportant source of noninsurable risk that can also be usefulfor ruling out the possibility that it is purely selection biascommon to all overseas shareholders that is driving the results.Prior to the introduction of shares in England, individuals seek-ing to invest overseas had to join a regulatory company and tradeon their own account or in small partnerships. Those who pos-sessed such human capital were naturally already exposed toforeign opportunities and risks from changes in Crown foreignpolicy. In contrast, the introduction of shares allowed nonmer-chants to gain such exposure for the first time. To test whethershareholding aligned the incentives of disparate groups, Iestimate:

Rebeli ¼X

j

gj ~xij þ ~�Merchanti þ ~�OverseasInvi

þ �ðMerchanti �OverseasInviÞ þ ui;

ð6Þ

where l provides the differential effect of shareholding for thosewith mercantile endowments. If the results are purely the resultof selection bias common to all shareholders, for example,through unobserved risk preferences or preexisting ideologicalpreferences, then there should be no differences between mer-chant and non-merchant shareholders, and l should be 0.Instead, if as I have argued, merchants already enjoyed expo-sure to overseas risks and opportunities, whereas nonmerchantsmostly lacked exposure to overseas risks and opportunities inthe absence of shares, ~� should be positive and l should be off-setting and negative.


Dow




ber 2019

IV. Sampling and Data

To become a member of the House of Commons, an individualhad to be selected by one of 249 constituencies, including thechartered boroughs, the 59 counties, and the universities ofOxford and Cambridge. I estimate that 144,737 individuals, or2.9 percent of the population, had the vote in 1628 (see OnlineAppendix 2.2.2). By the Long Parliament, the electorate ‘‘reacheddown not only to the minor gentry and rich merchants, but toyeomen, craftsmen, shopkeepers in the majority of towns andall the counties’’ (Plumb 1969, p. 103). With the summoning ofa new Parliament, writs were sent to all enfranchised boroughsand counties calling for representatives to be sent to London. Tobe elected ‘‘knight of the shire,’’ representing one’s county, was animportant social distinction for the most prominent landowners.Normally, county representatives would be chosen at a meetingat the county seat on the next county day, with voices in favor of aparticular candidate counted and returned by a designated sher-iff. All English and Welsh county seats had the same franchise—any males possessing 40 shillings’ worth of freehold land wereentitled to vote. Borough franchises varied considerably, withthe vote restricted to property owners or freemen in some to en-compassing all adult male inhabitants in others (see OnlineAppendix 2.2.2). Boroughs often sent town officials or acceptednominations from local patrons. Though technically MPs wereentitled to compensation for their expenses and service, manyboroughs preferred to choose members who were willing to sup-port themselves.

Not surprisingly then, MPs were selected from the Englishelite. Online Appendix Table A2.2 compares the social classifica-tion of fathers of members of the Long Parliament and 1628Parliament with Lindert and Williamson’s (1982) estimates ofthe relative incomes and proportions of different social groupsin 1688. MPs disproportionately came from families of the nobil-ity, gentry, lawyers, Crown officeholders, and merchants. Theaverage MP’s father in 1628–1640 came from a class that hadan average yearly income 690 percent of the averageEnglishman in 1688, with the median MP’s father being a gen-tleman with around 410 percent of the average yearly income.These wealth differences, along with a greater familiarity withLondon and financial and political capital, may have caused theinformation costs of learning about shares to be less pronounced


Dow




ber 2019






among this elite group, and may explain why wealth endowmentsdo not appear to be correlated with the propensity to hold sharesamong MPs (please also see Online Appendix 2.2.1). This elitesample may therefore better approximate an environmentwhere the decision to hold shares was as good as random, condi-tional on observable wealth and geographical endowments, thanone might expect for the population more broadly.

To construct the data, I consulted biographies of eachmember of the Long Parliament (and where available, thefather of each member), drawing in particular from compilationsby Keeler (1954), Brunton and Pennington (1954), and theHistory of Parliament Trust. Online Appendix 2.1 provides de-tails, including an example of such an entry. I augmented thesedata with 505 members of the 1628 Parliament, producing a dataset derived from consulting more than 1,842 MP and parentalbiographies in total. I also created a GIS of the 1,235 addressesof the fathers of each member of the 1628 and 1642–1660Parliaments to assess how endowed locational characteristics in-fluence the propensity to invest among MPs (see Figure I). I alsodraw on corroborative out-of-sample evidence from a thin data seton the joint stock investment and birthdates of all 1,667 membersof the Parliaments of 1603–1628 and 89 members of the House ofLords during the Civil War.

Each individual and his father were matched with Rabb’s(1967) lists of all investors in overseas companies mentioned inthe founding charters, patent rolls, and subsequent transferbooks of the major overseas trading companies founded inEngland between 1575 and 1630. In total, Rabb provides namesof 6,336 investors mentioned during this period. These investorlists were further supplemented and extended to 1640, wherepossible, using biographical information and the charters of theProvidence Island and Connecticut River companies. OnlineAppendix 2.1.2 provides the preferred matching algorithm thatI use for identifying overseas investors, while Online AppendixTable A1.10 shows the robustness of my results to alternativecodings.

I also collected a set of controls for other endowments that theliterature has emphasized as influencing support for parliamen-tary supremacy. For example, religion has played an importantrole in the historiography of the Civil War. A combination of twoproxies can be used to capture the effect of Puritanism. First,biographical data identifies individuals who attended Puritan


Dow




ber 2019







seminaries or colleges that had strong Puritan ties. An MP’s ed-ucation at such institutions may be interpreted as an indicator ofPuritan preferences. To capture religious preferences amongthose who did not attend such institutions, I also gathered data,based on diocesan records, of active Puritan ministers andCatholic recusants in the county each MP represented in1600.18 Similarly, to gauge whether an individual’s family hadentered the gentry prior to the Tudor dynasty, I followed bio-graphical records of the ancestors of each MP as far back as theBattle of Bosworth Field in 1485, which led to their accession.

As discussed already, MPs were also selected by constitu-encies with varying franchises. I create estimates of the numberof voters in 1628, as well as indicator variables for franchisetype in that year (Online Appendix 2.2.2). I also coded whetherthere was an electoral contest during the Long Parliament(true of 23.5 percent of counties and 23.3 percent of boroughs)or in the period 1603–1640 (39.2 percent and 22.8 percent,respectively).19

I also examine a sample of particular interest: the set of con-stituencies that constituted the major tax base of England, andwhose members of Parliament were relatively free from Crowninfluence. A useful gauge of constituency wealth, population, andrelative independence comes from the extent to which constitu-encies paid taxes during the extensive Tudor lay subsidy surveyof 1524–1525 (Online Appendix 2.2.3). Constituencies that didnot appear as separate jurisdictions at the time of the lay subsidyinclude those rotten boroughs that simply lacked any taxpayersas well as certain frontier counties that had separate taxationarrangements and over which the Crown enjoyed greater discre-tion and influence (Online Appendix Figure A2.2). As a result,those constituencies that did appear in the subsidy arguably con-stituted the core constituencies of wealthholders during the CivilWar.

18. These were at the ancient diocese level (McGrath 1967). To get estimates atthe county level, a uniform distribution of ministers and recusants per unit of areain a diocese was assumed.

19. These contests could be close elections, but among borough constituenciescould also be the submission of double returns due to rival sets of individuals claim-ing the right to vote.


Dow




ber 2019




V. Results

As the descriptive statistics in Table I reveal, overseasshareholders constituted 17.2 percent of members of the LongParliament, or 91 members, according to my preferred match-ing algorithm (though this figure could have been as high as23.5 percent or 125 members; Online Appendix 2.1.2). As themeans comparisons suggest, consistent with the importance ofinformation costs in shaping financial investment, sons ofshareholders were more likely to invest. Consistent with thepresence of local biases in share investment, MPs whose fathershad lived closer to London were also more likely to invest.However, investment was not just a London story. Investor-MPs were distributed throughout the country, as far north asCarlisle and Berwick-upon-Tweed, and there is significant var-iation both between and within counties of origin in the pro-pensity to invest and to subsequently rebel (Figure I).Similarly, MPs from mercantile backgrounds constituted 17.4percent of the Long Parliament and 35.2 percent of overseasinvestors. I control for all these factors explicitly and showthe effect comes mainly from nonmerchants.

Apart from these differences, shareholder and nonshare-holder MPs seem strikingly similar along a range of other dimen-sions, including central determinants of wealth. Similarproportions of shareholder and nonshareholder MPs were heirs,inherited ties to the royal court, inherited manors or landed es-tates, or had fathers who were nobles, knights, or baronets. MPseducated in Puritan colleges or seminaries also do not seem tohave been more likely to invest.

Despite the broad similarities between the endowments ofshareholders and nonshareholders, there are large differencesin their political decisions. Of shareholder MPs, 75.8 percentsupported parliamentary supremacy in the Civil War, comparedwith 51.6 percent of nonshareholders. Shareholder MPs alsowere more likely to support Parliament against the Crown atother points in the life cycle of the struggle for which roll callevidence survives: they were three times as likely as nonshare-holders to support the conviction of the Charles’s chief advisor,the Earl of Strafford, in 1641, for having suggested the use ofIrish troops on quell mainland protest. Of shareholder MPs,59.3 percent offered a loan to defend Parliament in Londonon the eve of hostilities in 1642, compared with 34.0 percent


Dow




ber 2019


TA

BL

EI

SU

MM

AR

YS

TA

TIS

TIC

S,

BY

OV

ER

SE

AS

SH

AR

EH

OL

DIN

GA

ND

SU

PP

OR

TF

OR

PA

RL

IAM

EN

TA

RY

SU

PR

EM

AC

Y(R

EB

EL

LIO

N)

Ou

tcom

esL

ong

Parl

iam

ent

Roy

ali

sts

Reb

els

Non

inves

tors

OJS

inves

tors

NM

ean

Std

.d

ev.

NM

ean

Std

.d

ev.

NM

ean

Std

.d

ev.

Pro

b>jtj

NM

ean

Std

.d

ev.

NM

ean

Std

.d

ev.

Pro

b>jtj

Parl

iam

enta

ryre

bel

,1642–1648

529

0.5

58

0.4

97

234

0.0

00

295

1.0

00

438

0.5

16

0.5

00

91

0.7

58

0.4

31

0.0

00

Hel

dco

urt

/patr

onage

offi

cebef

ore

1640

529

0.3

01

0.4

59

234

0.4

10

0.4

93

295

0.2

14

0.4

10

0.0

00

438

0.2

74

0.0

21

91

0.4

29

0.4

98

0.0

03

Su

pp

orte

dC

row

nad

vis

or(S

traff

ord

)1640

529

0.1

23

0.3

29

234

0.2

61

0.4

40

295

0.0

14

0.1

15

0.0

00

438

0.1

39

0.3

47

91

0.0

44

0.2

06

0.0

12

Con

trib

ute

dto

Lon

don

’sD

efen

ce1642

529

0.3

84

0.4

87

234

0.1

15

0.3

20

295

0.5

97

0.4

91

0.0

00

438

0.3

40

0.4

74

91

0.5

93

0.4

94

0.0

00

Sat

inR

um

pP

arl

iam

ent

1648–1653

516

0.2

56

0.4

37

223

0.0

04

0.0

67

293

0.4

47

0.4

98

0.0

00

426

0.2

37

0.4

26

90

0.3

44

0.4

78

0.0

34

Log

inco

me,

ca.

1640–50

265

6.7

98

1.1

65

180

6.7

66

1.1

59

85

6.8

65

1.1

83

0.5

19

221

6.7

29

1.1

40

44

7.1

42

1.2

41

0.0

32

Vari

able

sof

inte

rest

Inves

tor

inov

erse

as

join

tst

ock

co.

529

0.1

72

0.3

78

234

0.0

94

0.2

92

295

0.2

34

0.4

24

0.0

00

438

0.0

00

91

1.0

00

Att

ain

edad

ult

hoo

din

over

seas

IPO

yea

r529

0.2

08

0.4

06

234

0.1

75

0.3

81

295

0.2

34

0.4

24

0.0

99

438

0.1

94

0.3

96

91

0.2

75

0.4

49

0.0

85

Ind

ivid

ual

end

owm

ent

con

trol

sF

ath

erin

ves

tor

inov

erse

as

join

tst

ock

529

0.1

93

0.3

95

234

0.1

97

0.3

98

295

0.1

90

0.3

93

0.8

45

438

0.1

71

0.3

77

91

0.2

97

0.4

59

0.0

06

Fath

erm

erch

an

tor

ap

pre

nti

ced

mer

chan

t529

0.1

74

0.3

79

234

0.1

07

0.3

10

295

0.2

27

0.4

20

0.0

00

438

0.1

37

0.3

44

91

0.3

52

0.4

80

0.0

00

Gen

tlem

an

(in

her

ited

am

an

or)

529

0.5

48

0.4

98

234

0.5

47

0.4

99

295

0.5

49

0.4

98

0.9

61

438

0.5

43

0.4

99

91

0.5

71

0.4

98

0.6

26

Inh

erit

edla

nd

529

0.7

56

0.4

30

234

0.7

69

0.4

22

295

0.7

46

0.4

36

0.5

33

438

0.7

60

0.4

27

91

0.7

36

0.4

43

0.6

28

Hei

r529

0.7

28

0.4

46

234

0.7

44

0.4

38

295

0.7

15

0.4

52

0.4

68

438

0.7

31

0.4

44

91

0.7

14

0.4

54

0.7

51

Fath

erk

nig

ht

orbaro

net

529

0.3

18

0.4

66

234

0.3

42

0.4

75

295

0.2

98

0.4

58

0.2

86

438

0.3

15

0.4

65

91

0.3

30

0.4

73

0.7

86

Fath

ern

oble

529

0.0

91

0.2

88

234

0.1

24

0.3

30

295

0.0

64

0.2

46

0.0

18

438

0.0

98

0.2

98

91

0.0

55

0.2

29

0.1

92

Exp

erie

nce

dw

ard

ship

529

0.0

98

0.2

98

234

0.1

11

0.3

15

295

0.0

88

0.2

84

0.3

79

438

0.0

96

0.2

95

91

0.1

10

0.3

14

0.6

84

Gen

try

pri

orto

the

Tu

dor

s529

0.3

76

0.4

85

234

0.4

27

0.4

96

295

0.3

36

0.4

73

0.0

31

438

0.3

86

0.4

87

91

0.3

30

0.4

73

0.3

15

Inh

erit

edco

urt

ties

529

0.3

91

0.4

89

234

0.4

87

0.5

01

295

0.3

15

0.4

65

0.0

00

438

0.3

86

0.4

87

91

0.4

18

0.4

96

0.5

73

Log

.m

in.

dis

t.fa

ther

’sre

sid

ence

toL

ond

on528

10.8

52

2.0

91

234

10.9

60

2.1

19

294

10.7

67

2.0

69

0.2

91

437

10.9

74

2.0

32

91

10.2

68

2.2

77

0.0

03

#F

ath

er’s

ad

dre

sses

529

1.5

35

0.8

32

234

1.5

94

0.8

80

295

1.4

88

0.7

90

0.1

46

438

1.5

27

0.8

33

91

1.5

71

0.8

32

0.6

46

Att

end

edP

uri

tan

sem

inary

529

0.1

44

0.3

51

234

0.1

03

0.3

04

295

0.1

76

0.3

82

0.0

16

438

0.1

44

0.3

51

91

0.1

43

0.3

52

0.9

81


Dow




ber 2019

FIG

UR

EI

Over

seas

Inves

tmen

tan

dC

ivil

War

Reb

elli

onby

Fath

er’s

Ad

dre

ss

MP

sw

ith

pare

nta

lad

dre

sses

acr

oss

the

cou

ntr

ybot

hin

ves

ted

an

dre

bel

led

.P

oole

dsa

mp

lein

clu

din

gall

MP

sth

at

sat

inth

e1628

(an

dw

ere

ali

ve

in1641)

an

dth

eL

ong

Parl

iam

ent.


Dow




ber 2019

of others.20 Of investor-MPs still alive, 34.4 percent sat in theRump Parliament of Civil War victors that implemented manyof England’s changes in foreign policy from 1648 to 1653, com-pared with 23.7 percent of noninvestors. Online AppendixFigure A1.3 breaks these proportions down by mercantile back-ground at critical moments in the struggle. Even in the rawproportions there appears to be an alignment effect of shares:nonmerchants were more likely to be royalists in the absence ofshares, whereas nonmerchant shareholders joined merchants inrebellion.21

Table I reveals important similarities and significant differ-ences between royalists and supporters of parliamentary suprem-acy in other dimensions as well. In particular, notice thatvirtually all measures of endowed wealth appear similar amongboth populations, suggesting that the Civil War was not fought toprotect domestic wealth.22

In addition, 42.7 percent of MPs from pre-Tudor gentry back-grounds were royalists, rather than 33.6 percent of other MPs.However, as I show later, although there is some evidence thatolder gentry appear to support the royalists relative to other MPs,new gentry—those who inherited manors from families that at-tained manorial status during or after the Tudor era—are not anymore likely to rebel than the rest of MPs. Thus differences thatseem to be consistent at first with the newly commercializedgentry hypothesis (e.g., Tawney 1941; Moore 1966; Rajan andZingales 2003) instead appear to reflect the relative conservatism

20. In June 1642, claiming that the ‘‘King (seduced by wicked Counsel) intendsto make War against his Parliament,’’ the Long Parliament passed an ordinancesoliciting a loan to ‘‘uphold the Power and Privileges of Parliament.’’ Lists survive ofthe subscribers who were promised the return of their funds with 8 percent interest,and that ‘‘no Man’s affection [to Parliament and its privileges] shall be measured bythe Proportion of his offer,’’ but by the act of participation (Firth and Rait 1911, pp.6–9).

21. However, these are not the ‘‘new merchants’’ credited by Brenner (1993) forsupporting reform, as these MPs were apprenticed in traditional merchant compa-nies or had fathers who were merchants, and thus were from more establishedmercantile lineages. In fact, the effect of old mercantile endowments, which findsvalidation across specifications (Table IV) appears inconsistent with Brenner’sview that established mercantile groups supported the king. As I argue, whilethese groups had definite interests in courting the king to protect their propertyrights under the existing regime, they also had incentives to reduce political risk bysupporting parliamentary supremacy.

22. There do appear to be some differences: of the 48 MPs who were children ofnobles, two-thirds supported the king. We will discuss this below.


Dow




ber 2019



of older gentry in supporting the Crown rather than the initiativeof newer gentry in supporting representative government to pro-tect newly acquired property rights.

The Reformation in the sixteenth century also brought theCrown broad constitutional discretion over the Anglican Churchand an expansion of the Crown bureaucracy and patronagesystem. Both religious endowments and the value of inheritedties to the court were deeply affected by regime change, and myframework implies that they should both have greater influenceon an MP’s propensity to support the monarchy or to rebel.23

These suggest useful falsification tests, jointly of the frameworkand the precision of the data, if such effects are lacking. As Table Ipreviews, these tests are rejected: MPs with inherited court tieswere significantly more likely to be royalists, whereas MPs thatattended Puritan seminaries were more likely to rebel.

V.A. Lower Bound Estimates of Financial Assetholding

Table II presents regression results on the propensity toinvest in overseas shares among MPs. In this and subsequenttables, I begin by providing a regression with a small numberof controls to help build intuition (column (1)). These includedummy variables for whether an MP’s father was an overseasinvestor, whether the MP was apprenticed to a merchant guildor was the son of a merchant, and whether the MP inherited orstood to inherit a manor—an indicator of gentry status. I nextmatch on other wealth, geographical, and other individual en-dowments (column (2)). I progressively match on constituencycharacteristics (column (3)), county of constituency, and franchisetype (column (4)) and restrict the sample to MPs representingcore constituencies that paid taxes in the Tudor lay wealth

23. Some see the so-called Puritan Revolution as a religious conflict, part of thebroader Wars of Religion (Morrill 1993). This interpretation appears incomplete.Many considered Puritans were labeled as such ex post, often for actions not clearlymotivated by endowed religious preferences. As the Grand Remonstrance suggests,reformers favored a ‘‘profitable’’ war in the Spanish West Indies instead of directconflict with Catholic Spain. Religion also was useful propaganda. As the LongParliament MP John Selden wrote: ‘‘the very Arcanum of pretending religion inall wars is that something may be found out in which all men may have interest. Inthis the groom has as much interest as the lord. Were it for land, one has one thou-sand acres and the other but one; he would not venture so far as he that has athousand. But religion is equal to both. Had all men land alike, then . . . all menwould say they fought for land’’ (Hill 1961, p. 105).


Dow




ber 2019

TA

BL

EII

DE

TE

RM

INA

NT

SO

FD

EC

ISIO

NT

OH

OL

DO

VE

RS

EA

SS

HA

RE

S(1

575–1640)

AM

ON

GM

EM

BE

RS

OF

PA

RL

IAM

EN

T

Reg

ress

ion

:in

ves

tor

inov

erse

as

join

tst

ock

(0/1

)(1

)(2

)(3

)(4

)(5

)(6

)(7

)(8

)(9

)P

robit

,d

F/d

XP

robit

,d

F/d

XP

robit

,d

F/d

XO

LS

OL

SO

LS

OL

SO

LS

OL

S

Fath

erin

ves

tor

inov

erse

as

join

tst

ock

0.0

79**

0.0

66

0.0

80*

0.0

71

0.0

55

0.0

55

0.0

67

0.0

54

0.0

45

[0.0

40]

[0.0

54]

[0.0

48]

[0.0

54]

[0.0

49]

[0.0

49]

[0.0

44]

[0.0

42]

[0.0

45]

Fath

erm

erch

an

tor

ap

pre

nti

ced

mer

chan

t0.2

13**

*0.2

04**

*0.1

72**

*0.1

55**

0.1

41

0.1

37

0.1

27**

0.1

08

0.1

40

[0.0

59]

[0.0

64]

[0.0

63]

[0.0

73]

[0.0

84]

[0.0

84]

[0.0

62]

[0.0

74]

[0.0

86]

Inh

erit

eda

man

or0.0

38

0.0

43

0.0

37

0.0

33

0.0

51

0.0

62

0.0

48

0.0

71

0.0

53

[0.0

29]

[0.0

46]

[0.0

43]

[0.0

39]

[0.0

41]

[0.0

41]

[0.0

37]

[0.0

43]

[0.0

42]

Inh

erit

edla

nd

�0.0

13

�0.0

15

0.0

04

�0.0

04

�0.0

26�

0.0

24

�0.0

43

�0.0

58

[0.0

53]

[0.0

51]

[0.0

48]

[0.0

50]

[0.0

47]

[0.0

43]

[0.0

47]

[0.0

49]

Hei

r�

0.0

18

�0.0

18

0.0

04

0.0

17

0.0

20�

0.0

07

0.0

06

0.0

28

[0.0

39]

[0.0

39]

[0.0

40]

[0.0

45]

[0.0

46]

[0.0

34]

[0.0

38]

[0.0

44]

Fath

erk

nig

ht

orbaro

net

0.0

00

�0.0

19

�0.0

18

�0.0

09

�0.0

06�

0.0

10

0.0

00

0.0

26

[0.0

40]

[0.0

37]

[0.0

39]

[0.0

44]

[0.0

46]

[0.0

34]

[0.0

44]

[0.0

47]

Fath

ern

oble

�0.0

87*

�0.0

96**

*�

0.0

76

�0.0

53

�0.0

50�

0.0

81

�0.0

44

�0.0

12

[0.0

48]

[0.0

33]

[0.0

57]

[0.0

61]

[0.0

61]

[0.0

55]

[0.0

65]

[0.0

65]

Exp

erie

nce

dw

ard

ship

0.0

23

0.0

33

0.0

56

0.0

29

0.0

36

0.0

31

0.0

06

0.0

44

[0.0

49]

[0.0

55]

[0.0

54]

[0.0

62]

[0.0

62]

[0.0

53]

[0.0

62]

[0.0

63]

Gen

try

pri

orto

the

Tu

dor

s0.0

16

0.0

09

0.0

18

0.0

22

0.0

20

0.0

23

0.0

28

0.0

14

[0.0

31]

[0.0

32]

[0.0

34]

[0.0

44]

[0.0

44]

[0.0

32]

[0.0

43]

[0.0

47]

Inh

erit

edco

urt

ties

0.0

32

0.0

53

0.0

43

0.0

37

0.0

34

0.0

28

0.0

14

0.0

42

[0.0

43]

[0.0

43]

[0.0

47]

[0.0

57]

[0.0

58]

[0.0

37]

[0.0

47]

[0.0

59]

Log

.m

in.

dis

t.fa

ther

’sre

sid

ence

toL

ond

on�

0.0

18**

�0.0

11

�0.0

17

�0.0

20

�0.0

16�

0.0

13

�0.0

09

�0.0

09

[0.0

09]

[0.0

09]

[0.0

12]

[0.0

16]

[0.0

16]

[0.0

10]

[0.0

14]

[0.0

14]

Att

end

edP

uri

tan

sem

inary

0.0

38

0.0

24

0.0

26

0.0

43

0.0

56

0.0

26

0.0

45

0.0

40

[0.0

64]

[0.0

47]

[0.0

56]

[0.0

62]

[0.0

62]

[0.0

53]

[0.0

62]

[0.0

66]


Dow




ber 2019

TA

BL

EII

(CO

NT

INU

ED)

Reg

ress

ion

:in

ves

tor

inov

erse

as

join

tst

ock

(0/1

)(1

)(2

)(3

)(4

)(5

)(6

)(7

)(8

)(9

)P

robit

,d

F/d

XP

robit

,d

F/d

XP

robit

,d

F/d

XO

LS

OL

SO

LS

OL

SO

LS

OL

S

Por

tco

nst

itu

ency

0.0

80*

0.1

09**

0.1

29*

0.0

99*

0.1

03*

0.1

01*

0.0

96

[0.0

45]

[0.0

50]

[0.0

63]

[0.0

56]

[0.0

53]

[0.0

56]

[0.0

57]

Log

.d

ista

nce

toco

nst

itu

ency

from

Lon

don

0.0

01

�0.1

98**

*�

0.1

82**�

0.2

00*�

0.2

01**

�0.2

10*

�0.2

16*

[0.0

31]

[0.0

71]

[0.0

88]

[0.1

03]

[0.0

77]

[0.1

13]

[0.1

25]

Pu

rita

nm

inis

ters

per

10,0

00

inco

un

ty,

1600

0.0

27

[0.0

34]

Log

.p

opu

lati

ond

ensi

tyin

cou

nty

0.1

25**

[0.0

62]

Log

.#

vot

ers

1628

0.0

12

�0.0

01

�0.0

14

�0.0

07�

0.0

14

�0.0

11

�0.0

18

[0.0

13]

[0.0

12]

[0.0

22]

[0.0

24]

[0.0

12]

[0.0

18]

[0.0

23]

Log

.la

ysu

bsi

dy

1524-2

5,

pd

s.0.0

03

�0.0

07

�0.0

05

�0.0

03

[0.0

20]

[0.0

20]

[0.0

17]

[0.0

20]

Sam

ple

LP

LP

LP

LP

LP

LP

LP

+1628

(ali

ve)

LP

+1628

(ali

ve)

LP

+1628

(part

isan

)

Ind

iv.

wea

lth

an

dot

her

end

owm

ent

con

trol

sN

oY

esY

esY

esY

esY

esY

esY

esY

esC

onst

itu

ency

con

trol

sN

oN

oY

esY

esY

esY

esY

esY

esY

esF

ran

chis

e(1

628)

FE

an

dco

un

tyof

con

st.

FE

No

No

No

Yes

Yes

Yes

Yes

Yes

Yes

Log

lay

subsi

dy

con

trol

an

dsu

b-s

am

ple

No

No

No

No

Yes

Yes

No

Yes

Yes

Om

itM

idd

lese

x?

No

No

No

No

No

Yes

No

Yes

Yes

Obse

rvati

ons

528

528

528

528

418

409

699

530

437

(Pse

ud

o)R

20.0

50.0

70.1

40.3

00.3

00.2

80.2

40.2

40.2

7

Not

es:

Th

ista

ble

pro

vid

esre

gre

ssio

nes

tim

ate

sof

the

det

erm

inan

tsof

the

dec

isio

nto

hol

dov

erse

as

share

sbet

wee

n1575

an

d1640

am

ong

MP

s.C

olu

mn

s(1

)–(6

)are

the

Lon

gP

arl

iam

ent

(LP

)sa

mp

le.

Col

um

ns

(7)–

(8)

poo

lM

Ps

from

the

1628

Parl

iam

ent

stil

lali

ve

in1642.

Col

um

n(9

)re

stri

cts

the

poo

led

1628

MP

sto

thos

ew

hos

eC

ivil

War

all

egia

nce

was

kn

own

.R

obu

stst

an

dard

erro

rscl

ust

ered

at

the

cou

nty

ofm

ost

rece

nt

con

stit

uen

cy.

*10%

,**

5%

,**

*1%

.In

div

idu

al

con

trol

sin

clu

de:

fath

erJS

inves

tor,

ap

pre

nti

ced

orfa

ther

mer

chan

t,gen

try

bef

ore

Tu

dor

s,in

her

ited

cou

rtti

es,

log.

min

imu

md

ista

nce

offa

ther

’sad

dre

ssto

Lon

don

,n

um

ber

offa

ther

’sad

dre

sses

,att

end

eda

Pu

rita

nse

min

ary

.W

ealt

hen

dow

men

tsin

clu

de:

inh

erit

eda

man

or,

inh

erit

edla

nd

,h

eir,

fath

erk

nig

ht

orbaro

net

,fa

ther

nob

lean

dex

per

ien

ced

ward

ship

.C

onst

itu

ency

con

trol

sad

d:

Pu

rita

nm

inis

ters

per

cap

ita

inco

un

ty1600,

Cath

olic

recu

san

tsp

erca

pit

ain

cou

nty

1600,

bor

ough

,p

ort,

log.

pop

ula

tion

den

sity

,co

nst

itu

ency

inR

oyal

Dem

esn

e1415,

Con

stit

uen

cyco

nta

ins

cast

le1415,

log

dis

tan

ce:

con

stit

uen

cyto

Lon

don

,co

nst

itu

ency

exp

erie

nce

dco

nte

stin

LP

elec

tion

,co

nst

itu

ency

exp

erie

nce

dco

nte

stin

1603–1628

elec

tion

s,lo

g.

#vot

ers

1628

(est

d).

Col

um

ns

(5)–

(6),

(8)–

(9)

ad

da

con

trol

for

the

log.

lay

subsi

dy

wea

lth

ass

essm

ent

ofth

eco

nst

itu

ency

in1524–1525,

an

dre

stri

cts

the

sam

ple

toa

core

set

ofta

xp

ayin

gco

nst

itu

enci

es.

Col

um

ns

(6)

an

d(9

)d

rop

the

cou

nty

ofM

idd

lese

x,

wh

ich

incl

ud

esL

ond

on.


Dow




ber 2019

assessment (1524–1525), upon which I also match (column (5)).I also drop MPs that represent constituencies within MiddlesexCounty, the environs of London (column (6)). I pool the data toinclude MPs in both the Long Parliament and those still alivefrom the 1628 Parliament (column (7)), again restricting tothe core constituency sample (column (8)). Unlike the LongParliament MPs, members of the 1628 Parliament could sit onthe fence and avoid choosing a side, so for comparability I alsorestrict the pooled MPs to those for whom a Civil War allegiancewas known (column (9)).

As Table II suggests, sons of shareholders and those withmercantile endowments are more likely to invest. However,apart from some nonrobust evidence for a relative lack of invest-ment among the children of nobles, a range of measures of indi-vidual domestic wealth endowments appear to have little effecton an MP’s propensity to hold shares.24 Religious endowmentsalso do not influence the propensity to hold shares. Nor do inves-tor MPs come from constituencies with higher numbers of votersor greater wealth, as measured in the lay subsidy.25

In contrast, the raw mean differences of 24.2 percentagepoints in support for parliamentary supremacy between overseasshareholders and other MPs (Table I) appears to be robustto matching MPs over a range of endowed characteristics(Table III). Comparing MPs with similar gentry status, parentalmercantile endowments, and parental share ownership, overseasshare ownership raises an MP’s probability of supporting parlia-mentary supremacy over monarchical dictatorship during theCivil War by around 21.7 percentage points (column (1)).Matching MPs across a range of wealth and geographical endow-ments has little effect on this coefficient (column (2)). It could be

24. Online Appendix Table A.11 provides mean comparisons among members ofthe House of Lords in 1642 and shows that, 48 percent of overseas shareholdersrebelled, a proportion significantly higher than the 29.7 percent levels amongnonshareholders. Further, 56.3 percent of those nobles that came of age in anIPO year rebelled, relative to 30.1 percent in other years.

25. Investor MPs do appear to be more likely to come from port constituencies,constituencies closer to London, and from counties with greater population density.These correlations are again consistent with the literature on local biases of inves-tors (Coval and Moskowitz 1999; Zhu 2002): that shareholding occurred amonggroups more familiar with opportunities overseas, not just in London but acrossthe country. I control for these factors explicitly and show that they do not appear tochange the effect of shareholding on an MP’s propensity to rebel and do not appearthemselves to have a direct effect on propensity to rebel (Table III).


Dow




ber 2019


TA

BL

EII

I

DE

TE

RM

INA

NT

SO

FD

EC

ISIO

NT

OR

EB

EL

(1642–1648)

AM

ON

GM

EM

BE

RS

OF

PA

RL

IAM

EN

T

Reg

ress

ion

:p

arl

iam

enta

ryre

bel

(0/1

)

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

Pro

bit

,d

F/d

XP

robit

,d

F/d

XP

robit

,d

F/d

XO

LS

OL

SO

LS

OL

SO

LS

OL

S

Inves

tor

inov

erse

as

join

tst

ock

0.2

17**

*0.2

15**

*0.1

85**

*0.1

86**

*0.2

18**

*0.2

29**

*0.1

24**

0.1

42**

0.2

25**

*[0

.050]

[0.0

55]

[0.0

56]

[0.0

62]

[0.0

73]

[0.0

77]

[0.0

48]

[0.0

61]

[0.0

75]

Fath

erin

ves

tor

inov

erse

as

join

tst

ock

�0.0

63

�0.0

20

0.0

22

�0.0

25

�0.0

07

�0.0

04

�0.0

59�

0.0

51

�0.0

34

[0.0

53]

[0.0

77]

[0.0

71]

[0.0

70]

[0.0

81]

[0.0

82]

[0.0

61]

[0.0

64]

[0.0

74]

Fath

erm

erch

an

tor

ap

pre

nti

ced

mer

chan

t0.1

75**

0.0

97

0.0

84

0.0

65

0.0

87

0.0

88

0.0

88

0.1

41*

0.0

94

[0.0

72]

[0.0

84]

[0.0

83]

[0.0

82]

[0.0

87]

[0.0

89]

[0.0

65]

[0.0

72]

[0.0

87]

Inh

erit

eda

man

or0.0

26

0.0

51

0.0

30

0.0

48

0.0

19

0.0

20

0.0

55

0.0

21

0.0

06

[0.0

43]

[0.0

54]

[0.0

57]

[0.0

60]

[0.0

63]

[0.0

64]

[0.0

63]

[0.0

75]

[0.0

66]

Inh

erit

edla

nd

0.0

54

0.0

70

0.0

50

0.0

80

0.0

94

0.0

79

0.1

24

0.1

07

[0.0

81]

[0.0

82]

[0.0

93]

[0.0

96]

[0.0

99]

[0.0

77]

[0.0

98]

[0.0

98]

Hei

r�

0.0

83

�0.0

78

�0.0

89

�0.0

39

�0.0

45

�0.0

46

0.0

15

�0.0

62

[0.0

55]

[0.0

56]

[0.0

55]

[0.0

63]

[0.0

64]

[0.0

55]

[0.0

61]

[0.0

66]

Fath

erk

nig

ht

orbaro

net

�0.0

24

�0.0

52

�0.0

56

�0.0

45

�0.0

50

�0.0

65�

0.0

66

�0.0

41

[0.0

55]

[0.0

58]

[0.0

54]

[0.0

64]

[0.0

65]

[0.0

51]

[0.0

57]

[0.0

61]

Fath

ern

oble

�0.1

37

�0.1

54*�

0.1

15

�0.1

23

�0.1

25

�0.0

35�

0.0

41

�0.1

02

[0.0

95]

[0.0

93]

[0.1

09]

[0.1

14]

[0.1

15]

[0.0

69]

[0.0

79]

[0.1

04]

Exp

erie

nce

dw

ard

ship

�0.0

75

�0.0

80

�0.0

29

�0.0

17

�0.0

22

�0.0

28�

0.0

09

�0.0

36

[0.0

60]

[0.0

75]

[0.0

73]

[0.0

99]

[0.0

98]

[0.0

66]

[0.0

85]

[0.0

99]

Gen

try

pri

orto

the

Tu

dor

s�

0.0

57

�0.0

52

�0.0

57

�0.0

82*�

0.0

83*�

0.0

49�

0.0

90**

�0.0

63

[0.0

40]

[0.0

41]

[0.0

42]

[0.0

48]

[0.0

48]

[0.0

38]

[0.0

39]

[0.0

48]

Inh

erit

edco

urt

ties

�0.1

85**

*�

0.1

86**

*�

0.1

55**

*�

0.1

90**

*�

0.1

97**

*�

0.1

10**�

0.1

43**

*�

0.1

83**

*[0

.044]

[0.0

49]

[0.0

50]

[0.0

58]

[0.0

59]

[0.0

43]

[0.0

52]

[0.0

57]

Log

.m

in.

dis

t.fa

ther

’sre

sid

ence

toL

ond

on�

0.0

36**�

0.0

12

�0.0

17

�0.0

05

�0.0

09

�0.0

09

0.0

02

�0.0

11

[0.0

15]

[0.0

16]

[0.0

17]

[0.0

19]

[0.0

19]

[0.0

14]

[0.0

14]

[0.0

19]

Att

end

edP

uri

tan

sem

inary

0.1

79**

0.1

52*

0.1

27

0.0

93

0.0

81

0.1

48*

0.1

18

0.0

74

[0.0

84]

[0.0

81]

[0.0

91]

[0.0

88]

[0.0

89]

[0.0

77]

[0.0

81]

[0.0

86]

Por

tco

nst

itu

ency

�0.0

39

�0.0

14

�0.0

41

�0.0

22

0.0

29

0.0

26

�0.0

11

[0.0

58]

[0.0

74]

[0.0

89]

[0.0

88]

[0.0

64]

[0.0

72]

[0.0

79]


Dow




ber 2019

TA

BL

EII

I

(CO

NT

INU

ED)

Reg

ress

ion

:p

arl

iam

enta

ryre

bel

(0/1

)

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

Pro

bit

,d

F/d

XP

robit

,d

F/d

XP

robit

,d

F/d

XO

LS

OL

SO

LS

OL

SO

LS

OL

S

Pu

rita

nm

inis

ters

per

10,0

00

inco

un

ty,

1600

0.0

97**

*[0

.036]

Log

.d

ista

nce

toco

nst

itu

ency

from

Lon

don

�0.0

99**

0.1

04

0.1

29

0.1

39

�0.0

02

0.0

19

0.1

30

[0.0

41]

[0.0

73]

[0.0

98]

[0.1

17]

[0.0

73]

[0.1

03]

[0.1

14]

Log

.la

ysu

bsi

dy

1524-2

5,

pd

s.0.0

33

0.0

39*

0.0

07

0.0

24

[0.0

21]

[0.0

22]

[0.0

21]

[0.0

21]

[�2]F

-tes

t(�

wea

lth

end

owm

ents

=0)

5.9

85.3

40.7

80.3

40.4

11.1

91.2

90.4

0P

rob>

[�2]F

0.4

30.5

00.5

90.9

10.8

70.3

30.2

90.8

7

Sam

ple

LP

LP

LP

LP

LP

LP

LP

+1628

(ali

ve)

LP

+1628

(ali

ve)

LP

+1628

(part

isan

)

Ind

iv.

wea

lth

an

dot

her

end

owm

ent

con

trol

sN

oY

esY

esY

esY

esY

esY

esY

esY

esC

onst

itu

ency

con

trol

sN

oN

oY

esY

esY

esY

esY

esY

esY

esF

ran

chis

e(1

628)

FE

an

dco

un

tyof

con

st.

FE

No

No

No

Yes

Yes

Yes

Yes

Yes

Yes

Log

lay

subsi

dy

con

trol

an

dsu

bsa

mp

leN

oN

oN

oN

oY

esY

esN

oY

esY

esO

mit

Mid

dle

sex?

No

No

No

No

No

Yes

No

Yes

Yes

Obse

rvati

ons

528

528

528

528

418

409

699

530

437

(Pse

ud

o)R

20.0

40.0

80.1

40.3

00.3

10.3

00.2

60.2

70.2

8L

ower

bou

nd

trea

tmen

tef

fect

(Ost

er2014)

0.0

50.1

70.2

00.1

24

<0.2

9>

0.1

42

<0.3

5>

0.2

25

<0.6

3>

Not

es:

Th

ista

ble

pre

sen

tsre

gre

ssio

nes

tim

ate

sof

the

effe

cts

ofov

erse

as

share

hol

din

gon

the

dec

isio

nto

rebel

am

ong

MP

s.C

olu

mn

s(1

)–(6

)are

the

Lon

gP

arl

iam

ent

(LP

)sa

mp

le:

ad

ecis

ion

not

tore

bel

imp

lies

bei

ng

aro

yali

st.

Col

um

ns

(7)–

(8)

poo

lsM

Ps

from

the

1628

Parl

iam

ent

stil

lali

ve

in1642:

for

thes

e1628

MP

s,re

bel

=0

ind

icate

s‘‘d

idn

otre

bel

.’’C

olu

mn

(9)

rest

rict

sth

ep

oole

d1628

MP

sto

thos

ew

hos

eC

ivil

War

all

egia

nce

was

kn

own

:re

bel

=0

again

ind

icate

sro

yali

st.

Rob

ust

stan

dard

erro

rscl

ust

ered

at

the

cou

nty

ofm

ost

rece

nt

con

stit

uen

cy.

*10%

,**

5%

,**

*1%

.In

div

idu

al

con

trol

sin

clu

de

fath

erJS

inves

tor,

ap

pre

nti

ced

orfa

ther

mer

chan

t,gen

try

bef

ore

Tu

dor

s,in

her

ited

cou

rtti

es,

log.

min

imu

md

ista

nce

offa

ther

’sad

dre

ssto

Lon

don

,n

um

ber

offa

ther

’sad

dre

sses

,att

end

eda

Pu

rita

nse

min

ary

.W

ealt

hen

dow

men

tsin

clu

de

inh

erit

eda

man

or,

inh

erit

edla

nd

,h

eir,

fath

erk

nig

ht

orbaro

net

,fa

ther

nob

lean

dex

per

ien

ced

ward

ship

.C

onst

itu

ency

con

trol

sad

dP

uri

tan

min

iste

rsp

erca

pit

ain

cou

nty

1600,

Cath

olic

recu

san

tsp

erca

pit

ain

cou

nty

1600,

bor

ough

,p

ort,

log.

pop

ula

tion

den

sity

,co

nst

itu

ency

inR

oyal

Dem

esn

e1415,

con

stit

uen

cyco

nta

ins

cast

le1415,

log

dis

tan

ce:

con

stit

uen

cyto

Lon

don

,co

nst

itu

ency

exp

erie

nce

dco

nte

stin

LP

elec

tion

,co

nst

itu

ency

exp

erie

nce

dco

nte

stin

1603-1

628

elec

tion

s,lo

g.

#vot

ers

1628

(est

d).

Col

um

ns

(5)–

(6),

(8)–

(9)

ad

da

con

trol

for

the

log.

lay

subsi

dy

wea

lth

ass

essm

ent

ofth

eco

nst

itu

ency

in1524–1525,

an

dre

stri

cts

the

sam

ple

toa

core

set

ofta

xp

ayin

gco

nst

itu

enci

es.

Col

um

ns

(6)

an

d(9

)d

rop

the

cou

nty

ofM

idd

lese

x,

wh

ich

incl

ud

esL

ond

on.

Th

elo

wer

bou

nd

trea

tmen

tef

fect

use

sm

eth

odin

Ost

er(2

014),

ass

um

ingd

=1,

max

R2

=1.

Obse

rvable

con

trol

sst

ren

gth

enco

effi

cien

tin

colu

mn

s(7

)–(9

),so

the

OL

Sco

effi

cien

tis

the

imp

lied

low

erbou

nd

(th

ees

tim

ate

dad

just

edtr

eatm

ent

effe

ctis

rep

orte

din

an

gle

bra

cket

s).


Dow




ber 2019

that MPs’ support for parliamentary supremacy was shaped moreby the interests of their constituency than by their individualinvestments in shares. However, the effect of overseas sharehold-ing is robust to adding a range of controls for constituencies ofrepresentation (column (3)). Comparing MPs representing con-stituencies from within the same historic county, with similarfranchise rights and historic wealth levels among core taxpayingconstituencies actually strengthens the effect from 18.6 to 22.9percentage points (columns (4)–(6)). The effects of shareholdingdo diminish in magnitude (to 12.5 percentage points) when I in-clude the MPs of the 1628 Parliament who were alive during theLong Parliament (columns (7)–(8)), though this includes mem-bers that could choose to sit on the fence. Restricting the dataonly to those MPs for whom the decision not to rebel also impliesroyalist allegiance yields an effect of 22.5 percentage points, re-markably similar to the Long Parliament sample (column (9)).

In contrast, a range of measures of endowments of domesticwealth appear to have little effect on support for parliamentarysupremacy, including the MP’s status as an heir or the inheri-tance of a manorial estate that indicates membership in thegentry. Other wealth endowment measures appear to have noeffect, either individually or in a joint F-test. As the theoreticalframework suggests, the lack of any effect of domestic wealthendowment measures, despite the fact that these were strongdeterminants of contemporary income (Online Appendix TableA1.3), and were likely to influence large proportions of an indi-vidual’s asset portfolio, suggests that individuals’ expected fallsin expropriation risk that accompanied parliamentary supremacyto be almost precisely zero. Support for parliamentary supremacyappears unaffected by endowments of wealth, and thus it is un-likely the Civil War was fought primarily to defend domesticproperty.

As I have shown, the effect of overseas shareholding appearslargely invariant to adding wealth and geographical controlschosen to address the main sources of selection suggested bythe literature on the stockholding puzzle. However there stillmight be an unobserved driver of share ownership, such as a dif-ferent ideological view of the world, that might affect both shareownership and the propensity to support parliamentary suprem-acy. Furthermore, as Oster (2014) suggests, coefficient stabilitymay reflect robustness or may be a result of the lack of explana-tory power of observables. The bottom panel of Table III estimates


Dow




ber 2019



the lower bound treatment effect using her approach, allowing allthe variation in rebellion to be explained (i.e., max R2 = 1) andthat the unobservable of concern—such as ideology—to havethe same level of explanatory power as my expanding set of con-trols in columns (4)–(9). Notice that since the coefficient on share-holding actually strengthens with controls, and the controls raisethe explanatory power of the regression to around 30 percent ofthe variation in rebellion, the implied lower bound estimate ofaround 20 percentage points is very close to the conventional es-timates of around 22.9 percent, particularly in the preferred laysubsidy sample of core constituencies (column (6)). For simi-lar reasons, in the pooled regressions, the bias-adjusted treat-ment effect actually exceeds the conventional estimate (columns(7)–(9)).

Online Appendix Figure A1.7 provides a complementary ap-proach to assessing the sensitivity of the results to selection bias,following Imbens (2003). The figure compares the partial corre-lations between the observed covariates, share ownership, andsupport for Parliament, relative to the thresholds necessary foran omitted binomial variable to reduce the effect of shareholdingto insignificance at conventional levels. As the figure reveals, noobservable covariate is sufficiently correlated with either shareownership or support for Parliament to eliminate the effect ofshares at the conventional 5 percent significance level. Even atthe more stringent 1 percent level, only the positive correlationbetween shareholding and inherited court connections (whichwould conversely imply stronger support for the Crown) isstrong enough to have that effect.

It might appear from Table III that mercantile endow-ments also played little robust role in shaping support for par-liament in the Civil War. Table IV estimates equation 6, addingan interaction term between shareholding and mercantile en-dowments to the specifications in Table III. Although the pointestimates on shareholding show a slight increase with the ad-dition of this interaction term, there is also a robust, stronglysignificant, and offsetting negative interaction effect betweenshareholding and existing mercantile interests. These resultsare consistent with an alignment effect: by providing nonmer-chants with the opportunity to benefit from overseas trade andexpansion, the effect of shares on support for parliamentarycontrol should be greater relative to those with established mer-cantile endowments, who already enjoyed such opportunities


Dow




ber 2019


TA

BL

EIV

DIF

FE

RE

NT

IAL

EF

FE

CT

SO

FH

OL

DIN

GO

VE

RS

EA

SS

HA

RE

SO

NT

HE

DE

CIS

ION

TO

RE

BE

LA

MO

NG

MP

SW

ITH

AN

DW

ITH

OU

TM

ER

CA

NT

ILE

EN

DO

WM

EN

TS

Reg

ress

ion

(OL

S):

parl

iam

enta

ryre

bel

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

Inves

tor

inov

erse

as

join

tst

ock

0.3

02**

*0.2

98**

*0.2

43**

*0.2

75**

*0.3

10**

*0.3

16**

*0.1

65**

*0.1

89**

*0.3

15**

*[0

.049]

[0.0

53]

[0.0

60]

[0.0

65]

[0.0

76]

[0.0

77]

[0.0

52]

[0.0

63]

[0.0

74]

Inves

tor�

ap

pre

nti

ced

orfa

ther

mer

chan

t-0

.307**

*-0

.329**

*-0

.301**

*-0

.323**

*-0

.334**

*-0

.327**

-0.1

58

-0.1

89

-0.3

32**

[0.1

05]

[0.1

02]

[0.1

10]

[0.1

15]

[0.1

22]

[0.1

36]

[0.1

13]

[0.1

36]

[0.1

21]

Ap

pre

nti

ced

orfa

ther

mer

chan

t0.2

60**

*0.1

80**

*0.1

59**

0.1

51**

0.1

77**

0.1

76**

0.1

28**

0.1

90**

*0.1

84**

[0.0

57]

[0.0

64]

[0.0

60]

[0.0

66]

[0.0

66]

[0.0

66]

[0.0

61]

[0.0

65]

[0.0

69]

Inh

erit

eda

man

or0.0

21

0.0

43

0.0

23

0.0

54

0.0

26

0.0

25

0.0

57

0.0

23

0.0

09

[0.0

41]

[0.0

51]

[0.0

52]

[0.0

62]

[0.0

65]

[0.0

67]

[0.0

64]

[0.0

76]

[0.0

67]

Fath

erin

ves

tor

inov

erse

as

join

tst

ock

-0.0

60

-0.0

19

0.0

12

-0.0

26

-0.0

09

-0.0

09

-0.0

58

-0.0

54

-0.0

36

[0.0

50]

[0.0

71]

[0.0

64]

[0.0

71]

[0.0

82]

[0.0

83]

[0.0

62]

[0.0

64]

[0.0

76]

Joi

nt

F-t

est�(o

ver

seas

join

tst

ock

vari

able

s)=

019.6

716.7

88.2

99.5

78.6

28.8

5.1

24.4

79.5

8

Pro

b>

F0.0

00.0

00.0

00.0

00.0

00.0

00.0

10.0

20.0

0S

am

ple

LP

LP

LP

LP

LP

LP

LP

+1628

(ali

ve)

LP

+1628

(ali

ve)

LP

+1628

(part

isan

)


Dow




ber 2019

TA

BL

EIV

(CO

NT

INU

ED)

Reg

ress

ion

(OL

S):

parl

iam

enta

ryre

bel

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

Ind

iv.

wea

lth

an

dot

her

end

owm

ent

con

trol

sN

oY

esY

esY

esY

esY

esY

esY

esY

es

Con

stit

uen

cyco

ntr

ols

No

No

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Fra

nch

ise

(1628)

FE

an

dco

un

tyof

con

st.

FE

No

No

No

Yes

Yes

Yes

Yes

Yes

Yes

Log

lay

subsi

dy

con

trol

an

dsu

bsa

mp

leN

oN

oN

oN

oY

esY

esN

oY

esY

esO

mit

Mid

dle

sex?

No

No

No

No

No

Yes

No

Yes

Yes

Obse

rvati

ons

528

528

528

528

418

409

699

530

437

R2

0.0

60.1

20.1

80.3

10.3

20.3

10.2

60.2

70.3

Not

es:

Th

ista

ble

pro

vid

esO

LS

esti

mate

sof

the

effe

cts

ofov

erse

as

share

hol

din

gon

the

dec

isio

nto

rebel

am

ong

MP

s.It

incl

ud

esan

inte

ract

ion

term

bet

wee

nov

erse

as

inves

tor

an

dw

het

her

an

MP

was

ap

pre

nti

ced

am

erch

an

tor

wh

ose

fath

erw

as

am

erch

an

t.C

olu

mn

s(1

)–(6

)are

the

Lon

gP

arl

iam

ent

(LP

)sa

mp

le:

ad

ecis

ion

not

tore

bel

imp

lies

bei

ng

aro

yali

st.

Col

um

ns

(7)–

(8)

poo

lsM

Ps

from

the

1628

Parl

iam

ent

stil

lali

ve

in1642:

for

thes

e1628

MP

s,re

bel

=0

ind

icate

s‘‘d

idn

otre

bel

.’’C

olu

mn

(9)

rest

rict

sth

ep

oole

d1628

MP

sto

thos

ew

hos

eC

ivil

War

all

egia

nce

was

kn

own

:R

ebel

=0

again

ind

icate

sro

yali

st.

Rob

ust

stan

dard

erro

rscl

ust

ered

at

the

cou

nty

ofm

ost

rece

nt

con

stit

uen

cy.

*10%

,**

5%

,**

*1%

.In

div

idu

al

con

trol

sin

clu

de

fath

erJS

inves

tor,

ap

pre

nti

ced

orfa

ther

mer

chan

t,gen

try

bef

ore

Tu

dor

s,in

her

ited

cou

rtti

es,

log.

min

imu

md

ista

nce

offa

ther

’sad

dre

ssto

Lon

don

,n

um

ber

offa

ther

’sad

dre

sses

,att

end

eda

Pu

rita

nse

min

ary

.W

ealt

hen

dow

men

tsin

clu

de

inh

erit

eda

man

or,

inh

erit

edla

nd

,h

eir,

fath

erk

nig

ht

orbaro

net

,fa

ther

nob

lean

dex

per

ien

ced

ward

ship

.C

onst

itu

ency

con

trol

sad

dP

uri

tan

min

iste

rsp

erca

pit

ain

cou

nty

1600,

Cath

olic

recu

san

tsp

erca

pit

ain

cou

nty

1600,

bor

ough

,p

ort,

log.

pop

ula

tion

den

sity

,co

nst

itu

ency

inR

oyal

Dem

esn

e1415,

con

stit

uen

cyco

nta

ins

cast

le1415,

log

dis

tan

ce:

con

stit

uen

cyto

Lon

don

,co

nst

itu

ency

exp

erie

nce

dco

nte

stin

LP

elec

tion

,co

nst

itu

ency

exp

erie

nce

dco

nte

stin

1603–1628

elec

tion

s,lo

g.

#vot

ers

1628

(est

d).

Col

um

ns

(5)–

(6),

(8)–

(9)

ad

da

con

trol

for

the

log.

lay

subsi

dy

wea

lth

ass

essm

ent

ofth

eco

nst

itu

ency

in1524–1525,

an

dre

stri

cts

the

sam

ple

toa

core

set

ofta

xp

ayin

gco

nst

itu

enci

es.

Col

um

ns

(6)

an

d(9

)d

rop

the

cou

nty

ofM

idd

lese

x,

wh

ich

incl

ud

esL

ond

on.


Dow




ber 2019

and a relatively strong propensity to support Parliament (seealso Figure II).

Even though there is no significant effect of a range of mea-sures of domestic wealth on support for political reform, it stillmight be the case that insecure domestic property rights werecrucial in the decisions of agents to support political reform andthat the effect of shareholding occurs not through the alignmentof interests across groups in favor of control over sovereigntyrights but due to a desire to protect newly acquired wealth frominvestments in profitable overseas companies (as in Acemoglu,Johnson, and Robinson 2005). However the accumulation ofnew wealth does not appear to be driving these results. Table V(Panels A and B) compares shareholders who invested in overseascompanies that were unprofitable prior to the Civil War with

FIGURE II

Probability of Rebellion by Shareholding among Merchants and Nonmerchants

These boxplots depict the distribution of predicted probabilities of rebellionby overseas shareholding and mercantile endowments, based on a probit regres-sion of rebel (1) or royalist (0) on the set of individual endowment and constit-uency controls in Table III (column 2), (with the number of members in eachgroup in brackets). The central line depicts the median, the box depicts theinterquartile range. Notice that while individuals with mercantile endowmentswere highly likely to rebel regardless of shareholding, but there were only 97 ofthese in Parliament. Shareholding aligns the political allegiance of nonmerch-ants. In fact, the median nonmerchant who did not hold shares was actuallymore likely to be a royalist.


Dow




ber 2019

TA

BL

EV

DIF

FE

RE

NT

IAL

EF

FE

CT

SO

FH

OL

DIN

GS

HA

RE

SIN

UN

PR

OF

ITA

BL

EA

ND

CR

OW

N-P

RE

DA

TE

DO

VE

RS

EA

SC

OM

PA

NIE

SO

NT

HE

DE

CIS

ION

TO

RE

BE

LA

MO

NG

MP

S

OL

Sre

gre

ssio

n:

parl

iam

enta

ryre

bel

(0/1

)(1

)(2

)(3

)(4

)(5

)(6

)(7

)(8

)(9

)

Pan

elA

:U

np

rofi

table

JS

com

pan

ies

Inves

tor

inu

np

rofi

table

OJS

0.3

01**

*0.3

04**

*0.2

44**

*0.2

70**

*0.3

09**

*0.3

17**

*0.1

58**

*0.1

99**

*0.3

24**

*[0

.047]

[0.0

52]

[0.0

61]

[0.0

62]

[0.0

71]

[0.0

71]

[0.0

57]

[0.0

67]

[0.0

74]

Inves

tor

inu

np

rofi

table

OJS

xF

orA

.m

erch

an

t�

0.3

15**

*�

0.3

16**

*�

0.2

80**

*�

0.2

77**�

0.3

02**

*�

0.3

03**�

0.1

13

�0.1

52

�0.3

20**

*[0

.097]

[0.0

96]

[0.0

98]

[0.1

07]

[0.1

08]

[0.1

24]

[0.1

13]

[0.1

38]

[0.1

16]

Fath

erm

erch

an

tor

ap

pre

nti

ced

mer

chan

t0.2

60**

*0.1

68**

0.1

45**

0.1

33*

0.1

60**

0.1

62**

0.1

12*

0.1

72**

*0.1

70**

[0.0

60]

[0.0

69]

[0.0

64]

[0.0

72]

[0.0

71]

[0.0

71]

[0.0

59]

[0.0

61]

[0.0

72]

Joi

nt

FO

JS

inves

tor

vari

able

s21.3

117.4

98.3

49.7

610.1

010.6

33.8

04.3

610.2

8P

rob>

F0.0

00.0

00.0

00.0

00.0

00.0

00.0

30.0

20.0

0R

20.0

60.1

20.1

70.3

10.3

20.3

10.2

60.2

70.2

9

Pan

elB

:In

ves

tors

inu

np

rofi

table

com

pan

ies

rela

tive

toot

her

inves

tors

Inves

tor

inu

np

rofi

table

OJS

0.0

29

0.1

42

0.0

99

0.1

27

0.1

15

0.1

05

0.0

67

0.1

92

0.1

37

[0.1

86]

[0.1

57]

[0.1

63]

[0.1

88]

[0.2

27]

[0.2

77]

[0.1

66]

[0.2

17]

[0.1

86]

Inves

tor

inov

erse

as

join

tst

ock

0.1

87

0.0

72

0.0

69

0.0

69

0.1

10

0.1

31

0.0

64

�0.0

32

0.0

99

[0.1

80]

[0.1

56]

[0.1

61]

[0.1

98]

[0.2

37]

[0.2

84]

[0.1

54]

[0.2

07]

[0.1

90]

Joi

nt

FO

JS

inves

tor

vari

able

s9.7

78.3

25.1

75.3

74.9

24.8

83.4

53.0

94.8

Pro

b>

F0.0

00.0

00.0

10.0

10.0

10.0

10.0

40.0

60.0

1R

20.0

50.1

10.1

70.3

00.3

10.3

00.2

60.2

70.2

9P

an

elC

:In

ves

tors

inu

np

rofi

table

com

pan

ies

faci

ng

pre

dati

onby

the

Cro

wn

Inves

tor

inu

np

rofi

table

OJS

pre

date

dby

crow

n�

0.0

04

�0.0

20

�0.0

19

0.1

36

0.1

07

0.1

35

0.1

85

0.2

34

0.1

22

[0.1

14]

[0.1

24]

[0.1

26]

[0.1

64]

[0.1

83]

[0.1

98]

[0.1

36]

[0.1

73]

[0.1

86]

Inves

tor

inu

np

rofi

table

OJS

0.0

30.1

51

0.1

07

0.0

77

0.0

80

0.0

70

0.0

08

0.1

30

0.1

03

[0.1

83]

[0.1

60]

[0.1

65]

[0.1

91]

[0.2

23]

[0.2

70]

[0.1

74]

[0.2

19]

[0.1

78]

Inves

tor

inov

erse

as

join

tst

ock

0.1

87

0.0

72

0.0

68

0.0

68

0.1

05

0.1

17

0.0

59

�0.0

48

0.0

87

[0.1

80]

[0.1

56]

[0.1

61]

[0.1

98]

[0.2

37]

[0.2

88]

[0.1

54]

[0.2

06]

[0.1

93]

Joi

nt

FO

JS

inves

tor

vari

able

s6.9

46.0

33.7

53.7

23.7

93.7

62.4

02.0

53.9

6P

rob>

F0.0

00.0

00.0

20.0

20.0

20.0

20.0

80.1

30.0

2R

20.0

50.1

10.1

70.3

10.3

10.3

00.2

70.2

70.2

9


Dow




ber 2019

TA

BL

EV

(CO

NT

INU

ED)

OL

Sre

gre

ssio

n:

parl

iam

enta

ryre

bel

(0/1

)(1

)(2

)(3

)(4

)(5

)(6

)(7

)(8

)(9

)

Pan

elD

:E

ast

Ind

iaC

omp

an

yin

ves

tors

East

Ind

iaC

oin

ves

tor

�0.0

41

�0.1

22

�0.1

54

�0.1

75

�0.0

74

�0.0

45

�0.1

73

�0.1

19

�0.0

83

[0.1

49]

[0.1

39]

[0.1

52]

[0.1

56]

[0.1

47]

[0.1

68]

[0.1

16]

[0.1

63]

[0.1

52]

Inves

tor

inov

erse

as

join

tst

ock

0.2

22**

*0.2

28**

*0.1

90**

*0.2

21**

*0.2

32**

*0.2

37**

*0.1

62**

*0.1

64**

0.2

40**

*[0

.055]

[0.0

58]

[0.0

56]

[0.0

66]

[0.0

81]

[0.0

83]

[0.0

52]

[0.0

67]

[0.0

82]

Joi

nt

FO

JS

inves

tor

vari

able

s9.9

78.1

75.8

75.6

04.4

04.4

54.7

93.0

24.5

5P

rob>

F0.0

00.0

00.0

10.0

10.0

20.0

20.0

10.0

60.0

2R

20.0

50.1

10.1

70.3

10.3

10.3

00.2

70.2

70.2

9

Sam

ple

Lon

gP

arl

Lon

gP

arl

Lon

gP

arl

Lon

gP

arl

Lon

gP

arl

Lon

gP

arl

Lon

gP

arl

+1628

(ali

ve)

Lon

gP

arl

+1628

(ali

ve)

Lon

gP

arl

+1628

(part

isan

)

Ind

iv.

wea

lth

an

dot

her

end

owm

ent

con

trol

sN

oY

esY

esY

esY

esY

esY

esY

esY

esC

onst

itu

ency

con

trol

sN

oN

oY

esY

esY

esY

esY

esY

esY

esF

ran

chis

e(1

628)

FE

an

dco

un

tyof

con

st.

FE

No

No

No

Yes

Yes

Yes

Yes

Yes

Yes

Log

lay

subsi

dy

con

trol

an

dsu

bsa

mp

leN

oN

oN

oN

oY

esY

esN

oY

esY

esO

mit

Mid

dle

sex?

No

No

No

No

No

Yes

No

Yes

Yes

Obse

rvati

ons

528

528

528

528

418

409

699

530

437

Not

es:

Th

isre

gre

ssio

nta

ble

pro

vid

eses

tim

ate

sof

the

effe

cts

ofov

erse

as

share

hol

din

gon

the

dec

isio

nto

rebel

am

ong

MP

s.P

an

elA

incl

ud

esan

inte

ract

ion

term

bet

wee

nin

ves

tor

inan

un

pro

fita

ble

over

seas

com

pan

yan

dw

het

her

an

MP

was

ap

pre

nti

ced

am

erch

an

tor

wh

ose

fath

erw

as

am

erch

an

t.P

an

elB

incl

ud

esd

um

mie

sfo

rw

het

her

an

MP

inves

ted

inu

np

rofi

table

com

pan

ies

an

din

oth

erov

erse

as

com

pan

ies.

Pan

elC

ad

ds

afu

rth

erd

um

my

for

wh

eth

erth

eu

np

rofi

table

com

pan

yh

ad

exp

erie

nce

dC

row

np

red

ati

on.

Pan

elD

incl

ud

esa

du

mm

yfo

rw

het

her

the

MP

was

an

inves

tor

inth

eE

ast

Ind

iaC

omp

an

y.

Col

um

ns

(1)–

(6)

are

the

Lon

gP

arl

iam

ent

(LP

)sa

mp

le:

ad

ecis

ion

not

tore

bel

imp

lies

bei

ng

aro

yali

st.

Col

um

ns

(7)–

(8)

poo

lsM

Ps

from

the

1628

Parl

iam

ent

stil

lali

ve

in1642:

for

thes

e1628

MP

s,re

bel

=0

ind

icate

s‘‘d

idn

otre

bel

.’’C

olu

mn

(9)

rest

rict

sth

ep

oole

d1628

MP

sto

thos

ew

hos

eC

ivil

War

all

egia

nce

was

kn

own

:re

bel

=0

again

ind

icate

sro

yali

st.

Rob

ust

stan

dard

erro

rscl

ust

ered

at

the

cou

nty

ofm

ost

rece

nt

con

stit

uen

cy.

*10%

,**

5%

,**

*1%

.C

olu

mn

(1)

incl

ud

esco

ntr

ols

for

fath

erJS

inves

tor,

ap

pre

nti

ced

orfa

ther

mer

chan

tan

din

her

ited

am

an

or.

Col

um

n(2

)in

clu

des

ind

ivid

ua

lco

ntr

ols:

fath

erJS

inves

tor,

ap

pre

nti

ced

orfa

ther

mer

chan

t,gen

try

bef

ore

Tu

dor

s,in

her

ited

cou

rtti

es,

log.

min

imu

md

ista

nce

offa

ther

’sad

dre

ssto

Lon

don

,n

um

ber

offa

ther

’sad

dre

sses

,att

end

eda

Pu

rita

nse

min

ary

.W

ealt

hen

dow

men

tsin

clu

de

inh

erit

eda

man

or,

inh

erit

edla

nd

,h

eir,

fath

erk

nig

ht

orbaro

net

,fa

ther

nob

lean

dex

per

ien

ced

ward

ship

.C

onst

itu

ency

con

trol

sad

dP

uri

tan

min

iste

rsp

erca

pit

ain

cou

nty

1600,

Cath

olic

recu

san

tsp

erca

pit

ain

cou

nty

1600,

bor

ough

,p

ort,

log.

pop

ula

tion

den

sity

,co

nst

itu

ency

inR

oyal

Dem

esn

e1415,

con

stit

uen

cyco

nta

ins

cast

le1415,

log

dis

tan

ce:

con

stit

uen

cyto

Lon

don

,co

nst

itu

ency

exp

erie

nce

dco

nte

stin

LP

elec

tion

,co

nst

itu

ency

exp

erie

nce

dco

nte

stin

1603–1628

elec

tion

s,lo

g.

#vot

ers

1628

(est

d).

Col

um

ns

(5)–

(6),

(8)–

(9)

ad

da

con

trol

for

the

log.

lay

subsi

dy

wea

lth

ass

essm

ent

ofth

eco

nst

itu

ency

in1524–1525,

an

dre

stri

cts

the

sam

ple

toa

core

set

ofta

xp

ayin

gco

nst

itu

enci

es.

Col

um

ns

(6)

an

d(8

)–(9

)d

rop

the

cou

nty

ofM

idd

lese

x,

wh

ich

incl

ud

esL

ond

on.


Dow




ber 2019

otherwise similar MPs. The effect of shares on investors in un-profitable overseas joint stock companies appears to be strongerthan before—around 31.4 percentage points with full controls(column (6))—and once again, as the almost precisely offsettinginteraction term suggests, the effect stems from those withoutpreexisting mercantile interests.

Panel B adds controls for the main effect of overseas invest-ment to this regression. Adding up the point estimates in the laysubsidy sample with full controls (column (6)) reveals that thoughinvestors in profitable joint stock companies were around 13.1percentage points more likely to rebel, those in unprofitablejoint stock companies were around 23.6 percentage points morelikely to rebel than otherwise similar nonshareholder MPs.Though this result may appear somewhat counterintuitive, it isin fact consistent with the theoretical framework: those withprofitable overseas investments under the existing regime havesomething to lose (i.e., ��jjR is higher) with constitutional changerelative to those with nonperforming investments under the ex-isting regime.

Firms could be unprofitable for many reasons beyond politi-cal risk, such as bad luck or mismanagement. Panel C comparesthe propensities to rebel among those MPs that invested in sharesin companies that were both unprofitable and that had facedCrown predation (see Online Appendix Table A1.2). The pointestimates suggest investors in unprofitable companies predatedby the Crown were 32.2 percentage points more likely to rebelthan otherwise similar MPs. In contrast, though members of the(initially) profitable East India Company were around 19.2 per-centage points more likely than noninvestors to rebel, they were4.5 percentage points less likely than the average investor over-seas to do so (Panel D).26

It is possible to implement further tests of three key alterna-tive channels—unobserved components of domestic income, ide-ology and risk—that have been mooted as explanations for thesuccessful development of a coalition in favor of parliamentarysupremacy and which might also explain the effects of sharehold-ing. Online Appendix Table A1.3 tests whether these endowment

26. These results nuance Brenner’s (1993) claims that East India Companyinvestors were more likely to be royalists. Instead, East India investors inParliament were also more likely to rebel than noninvestors, but had lower propen-sities to rebel than other investors.


Dow




ber 2019



measures have any effect on that subsample for which measuresexist of a Long Parliament MP’s income at the time of the CivilWar. Notice that there are no robust differences in incomebetween shareholder and nonshareholder MPs. In contrast,Table A1.3 reveals that inheritances of landed estates and titlesare individually and jointly significant determinants of contem-poraneous income and alone explain 21 percent of the variation incontemporaneous income for those Long Parliament MPs forwhich contemporaneous income data survives (Table A1.3,column (1)). Furthermore, adding a richer set of constituencycontrols, franchise type, and county fixed effects explains up to67 percent of the variation in log income. These results suggeststhat my measures of endowed wealth do capture (and control for)important variation in contemporary income, and it is not impre-cision that is leading to a lack of a domestic wealth effect. As thetheoretical framework implies, the lack of effect of wealth endow-ments that also strongly predict contemporary income suggeststhat domestic assets were not perceived to be likely to change intheir expected returns (discounted by expropriation risk) withchanges in political regime.

Another possibility, as discussed already, is that the effectof shareholding is capturing unobserved preexisting differencesin views of the world or political allegiances. Such a preexistingideology story would suggest that in the years before the CivilWar, antimonarch shareholders would be also less likely toattend court or work for the Crown. In contrast, if the theoret-ical framework is correct, then prior to the stark decision to sup-port or oppose constitutional reform in the Long Parliament,individuals with endowments most subject to executive control,including shareholders, would paradoxically face a greater in-centive to secure their property by seeking favor and influenceat court.

Online Appendix Table A1.4 examines the effect of share-holding on the acquisition of royal office or court positions priorto the Long Parliament. Controlling for inherited court ties,nonmerchant shareholder MPs were between 8 and 14 percent-age points more likely to acquire court offices than other MPs(columns (1)–(9)). Thus, it appears that shareholders did attemptto work within the existing constitutional system to secure over-seas property: prior to the Long Parliament, shareholders werenot consistently opposed to the court, and many assumed courtroles.


Dow




ber 2019




Another alternative story might be that even controlling forwealth, overseas shareholder MPs are selected by risk prefer-ence, and even though both supporting the Crown and supportingParliament was inherently risky, they had lower status quo bias.One approach to addressing this is to check for differences inpolitical allegiance of investors in joint stock companies thatwere created to share risks, but did not face the political riskof overseas expropriation by the Crown. Online AppendixTable A1.5 examines the propensities to rebel among MPs whohad invested in joint stock companies involved in domestic ven-tures that were risky because of environmental risk rather thanCrown expropriation.27 As the table suggests, both with and with-out controls for overseas investment, there is no effect of joiningthese companies on the propensity to rebel.

Online Appendix Tables A1.6, A1.7, and A1.8 present resultsusing the same sets of controls to estimate the effect of shares onother indicators of support for parliamentary control of govern-ment over the life cycle of the struggle, including the legislativeattainder of the Earl of Strafford (1641), subscribing to theDefence of London on the eve of the war (1642) and membershipof the Rump Parliament of postwar victors (1648–1653). Thebroad picture that emerges is that at least as early as the peacefullegislative push for reform in 1641–1642, shareholders were con-sistently more likely than nonshareholders to oppose executiveauthority. This continued across the life cycle of the struggle.Furthermore, where merchants exhibit significant differencesto nonmerchants, shares appear to align nonmerchant choiceswith merchants. The fact that both nonmerchant shareholdersand merchants were significantly more likely to sit in the RumpParliament suggests a direct link between the genesis of the co-alition for reform and the dramatic changes in England’s publicinvestments in the navy and its foreign policy that would occur onits success.

27. I define ‘‘domestic’’ to be within the British Isles and its territorial waters.Examples of these companies include infrastructure companies such as the NewRiver Company; natural resource companies like the Mines, Minerals and BatteryWorks; fen drainage companies such as the Great Bedford Level; and companiesinvesting in the Fisheries, Ireland and wastelands in Wales.


Dow




ber 2019






V.B. Adulthood in an IPO Year as an Exogenous Driver ofShareholding

As in all quasi-experimental settings, the possibility thatsome residual unobserved selection process might bias theOLS results remains. However, it may also be the case thatthe OLS estimates, rather than being upper bounds, are actuallyunderestimates. This is due to the presence of a classic hold-upproblem. With the Crown able and willing to expropriate thereturns from overseas investments (Online Appendix TableA1.2), some individuals may have been motivated by the newaccess to overseas opportunities provided by shares to supportreform, who also faced incentives to wait until after the reform toactually invest. Indeed, a number of MPs who later gained prom-inence in the Civil War, most famously Oliver Cromwell, ac-tively considered but postponed joining ventures to the NewWorld. The omission of those motivated by these new opportu-nities would lead to a downward bias in the measured effect. Theresults could also be attenuated by the measurement error in-duced by using names and biographies to match individual MPsto joint stock companies.28

To assess whether this is the case, I exploit a series of shocksto the feasibility and enthusiasm to invest in overseas shares thattook place over this period, in combination with a feature ofEnglish common law that granted individuals the right to controltheir finances, sell their property, or write legally enforceablecontracts—including share contracts—at age 21 but not at age20. I compare individuals who invested in overseas shares be-cause they turned 21 in the year when an IPO of shares wasbeing made for a new overseas joint stock company to individualswho would have invested had they attained adulthood in such ayear. Some context may be useful for understanding the legalimplications of adulthood.

As discussed already, because share contracts lacked limitedliability and secondary markets, investors incurred both long-term financial and legal obligations. Thus the age at which indi-viduals became able to write legally enforceable contracts and

28. Online Appendix Table A1.10 shows that my main effects are robust to al-ternative codings of overseas joint stock investors, including a more restrictivedefinition based only on those explicitly identified as investors by the History ofParliament and Keeler (1954), and a broader definition that includes all feasiblename matches from the lists in Rabb (1967).


Dow




ber 2019




dispose of their inheritances—known as the age of majority—wasparticularly relevant for an individual’s ability to invest inshares. In his canonical Commentaries on the Laws of England,Blackstone (1769, p. 453) states: ‘‘It is generally true, that aninfant can neither aliene his lands, nor do any legal act, normake a deed, nor indeed any manner of contract, that will bindhim.’’ The full age of majority for both men and women was‘‘twenty one years, which age is completed on the day precedingthe anniversary of a person’s birth; who till that time is an infant,and so stiled in law’’ (Blackstone 1769, p. 451).29

MPs that sat in Parliaments from 1603 to 1628 and turned 21around Drake’s voyages and during years in which subsequentoverseas ventures issued stock were around 7 percentage pointsmore likely to invest in overseas ventures (relative to a baseline of15 percent) (Online Appendix Figure A1.5).30 Members of theLong Parliament naturally were younger at the time of Drake’svoyages, but those who came of age during subsequent IPO yearsalso showed spikes in their propensity to invest in overseas jointstock companies (Figure III). In fact, of the 110 Long ParliamentMPs that came of age in an IPO year, 22.7 percent invested inoverseas joint stock companies (compared with 15.8 percent ofother MPs) and 62.7 percent—or 69 MPs—rebelled in the CivilWar (compared with 53.9 percent). Reassuringly, individuals thatcame of age in IPO years also appear very similar along a range ofother endowed observable dimensions to those that came of age inother years (Table VI.)31

29. Though contracts were not legally binding below the age of 21, some indi-viduals, particularly sons of nobles and the children of some prominent merchants,were able to buy shares based on family or individual reputation even before theyturned 21. However, even for these categories of individuals, turning 21 in the yearof an overseas IPO tended to increase the propensities for investment (see OnlineAppendix Table A1.11 for members of the House of Lords and Figure V and TableVIII for those from mercantile backgrounds).

30. Of those still alive in 1641, a baseline of 13.0 percent invested while those ofage in an IPO year were 12.7 percentage points [4.50, 21.04] more likely to invest.

31. There are two exceptions: they appear more likely to have mercantile en-dowments and less likely to have fathers who were nobles. I will show that theresults are robust to controlling for these factors and that, once again, the effectis coming from nonmerchants.


Dow




ber 2019




I estimate regressions of the following form:

Rebeli ¼X

j

gjxij þ �OverInvi

þX

�¼�5��1;þ1�þ5

a� OfAge � yrsfromIPOþ ei

OverInvi ¼X

j

gjxij þ �0OfAgeIPOyri

þX

�¼�5��1;þ1�þ5

�� OfAge � yrsfromIPOþ i;

where OfAgeIPOyri indicates whether an individual turned 21in an IPO year, and in some specifications, I control for poten-tial cohort differences by including a set of dummy variables for

FIGURE III

Proportions Investing in Overseas Shares and Later Rebelling in Civil War byAge of Majority in Calendar Time: Long Parliament MPs

This graph presents the proportion of MPs that turned 21 in a particularyear who (A) invested in overseas joint stock companies prior to 1640 and (B)rebelled during the Civil War (1642–1648), drawing on MPs who sat in theLong Parliament, with the mean proportions smoothed locally across nearbyyears using weights from an epanechnikov kernel with bandwidth 0.3 (N = 533MPs). MPs who came of age just before or during IPO years for overseas jointstock companies (vertical dotted lines) often appear more likely to invest thatthose that come of age just after, and are often also more likely to rebel.


Dow




ber 2019

TA

BL

EV

I

ME

AN

CO

MP

AR

ISO

NS

BE

TW

EE

NM

PS

WH

OT

UR

NE

D21

INA

NIP

OY

EA

RA

ND

OT

HE

RL

ON

GP

AR

LIA

ME

NT

MP

S

Ou

tcom

esL

ong

Parl

iam

ent

Of

age

oth

eryea

rsO

fage

IPO

yea

r

NM

ean

Std

.d

ev.

NM

ean

Std

.d

ev.

NM

ean

Std

.d

ev.

Pro

b>jtj

Parl

iam

enta

ryre

bel

529

0.5

58

0.4

97

419

0.5

39

0.4

99

110

0.6

27

0.4

86

0.0

99

Log

inco

me,

ca1640–1650

265

6.7

98

1.1

65

211

6.8

69

1.1

35

54

6.5

21

1.2

49

0.0

50

Hel

dco

urt

/patr

onage

offi

cebef

ore

1640

529

0.3

01

0.4

59

419

0.2

94

0.4

56

110

0.3

27

0.4

71

0.4

93

Con

trib

ute

dto

Lon

don

’sD

efen

ce1642

529

0.3

84

0.4

87

419

0.3

79

0.4

86

110

0.4

00

0.4

92

0.6

94

Su

pp

orte

dC

row

nad

vis

or(S

traff

ord

)1640

529

0.1

23

0.3

29

419

0.1

15

0.3

19

110

0.1

55

0.3

63

0.2

56

Sat

inR

um

pP

arl

iam

ent

1648-1

653

516

0.2

56

0.4

37

406

0.2

41

0.4

28

110

0.3

09

0.4

64

0.1

49

Vari

able

ofin

tere

stIn

ves

tor

inov

erse

as

join

tst

ock

co529

0.1

72

0.3

78

419

0.1

58

0.3

65

110

0.2

27

0.4

21

0.0

85

Ind

ivid

ual

end

owm

ent

con

trol

sF

ath

erin

ves

tor

inov

erse

as

join

tst

ock

529

0.1

93

0.3

95

419

0.1

96

0.3

97

110

0.1

82

0.3

87

0.7

43

Fath

erm

erch

an

tor

ap

pre

nti

ced

mer

chan

t529

0.1

74

0.3

79

419

0.1

58

0.3

65

110

0.2

36

0.4

27

0.0

52

Gen

tlem

an

(in

her

ited

am

an

or)

529

0.5

48

0.4

98

419

0.5

47

0.4

98

110

0.5

55

0.4

99

0.8

81

Inh

erit

edla

nd

529

0.7

56

0.4

30

419

0.7

52

0.4

32

110

0.7

73

0.4

21

0.6

50

Hei

r529

0.7

28

0.4

46

419

0.7

40

0.4

39

110

0.6

82

0.4

68

0.2

24

Fath

erk

nig

ht

orbaro

net

529

0.3

18

0.4

66

419

0.3

32

0.4

71

110

0.2

64

0.4

43

0.1

73

Fath

ern

oble

529

0.0

91

0.2

88

419

0.1

05

0.3

07

110

0.0

36

0.1

88

0.0

26

Exp

erie

nce

dw

ard

ship

529

0.0

98

0.2

98

419

0.0

93

0.2

91

110

0.1

18

0.3

24

0.4

32

Gen

try

pri

orto

the

Tu

dor

s529

0.3

76

0.4

85

419

0.3

79

0.4

86

110

0.3

64

0.4

83

0.7

61

Inh

erit

edco

urt

ties

529

0.3

91

0.4

89

419

0.3

84

0.4

87

110

0.4

18

0.4

96

0.5

17

Log

.m

in.

dis

t.fa

ther

’sre

sid

ence

toL

ond

on528

10.8

52

2.0

91

418

10.8

81

2.0

71

110

10.7

44

2.1

74

0.5

41

#F

ath

er’s

ad

dre

sses

529

1.5

35

0.8

32

419

1.5

56

0.8

49

110

1.4

55

0.7

62

0.2

55

Att

end

edP

uri

tan

sem

inary

529

0.1

44

0.3

51

419

0.1

41

0.3

48

110

0.1

55

0.3

63

0.7

15


Dow




ber 2019

coming of age in each of the five years before or after an IPOyear, and a set of endowment controls xj.

Table VII (Panels A and B) shows the simplest reduced-formrelationships between investment and rebellion and whether anMP comes of age in an IPO year, over the Long Parliamentsample, the pooled sample with the 1628 Parliament andamong the core lay subsidy constituencies (columns (1)–(4)).The regressions leave out any controls other than an interceptfor being in the Long Parliament in the pooled specification.Observe that MPs that come of age in an overseas IPO yearare between 5.7 and 11.7 percentage points more likely toinvest in overseas joint stock companies (Panel A). They are be-tween 6.8 and 11.1 percentage points more likely to

TABLE VII

WALD IV ESTIMATES OF EFFECT OF SHAREHOLDING AMONG MPS ON DECISION TO REBEL

IN CIVIL WAR

(1) (2) (3) (4)Lay subsidy

constituencies

Panel A: OLS (first stage): overseas joint stock investorAttained adulthood in IPO year 0.071* 0.057* 0.117** 0.070*

[0.038] [0.030] [0.046] [0.035]F-test (excluded) 3.46 3.49 6.47 3.93Prob>F 0.07 0.07 0.02 0.06R2 0.01 0.02 0.01 0.02

Panel B: OLS (reduced form): parliamentary rebelAttained adulthood in IPO year 0.084 0.071** 0.111** 0.068**

[0.051] [0.032] [0.050] [0.033]R2 0.00 0.19 0.01 0.23

Panel C: IV-LIML (Wald estimator): parliamentary rebelInvestor in overseas joint stock 1.173 1.258* 0.954** 0.969**

[0.735] [0.693] [0.468] [0.476]Long Parliament intercept No Yes No YesSample LP LP + 1628 LP LP + 1628Observations 533 896 423 703

Notes: Panel A provides estimates of the first stage effect of turning 21 in an IPO year on an MP’sdecision to invest in overseas shares. Panel B provides the reduced-form effect of turning 21 in an IPOyear on a MP’s decision to rebel in the Civil War. Panel C provides the Wald IV estimates of effect onshareholding on the decision to rebel among MPs who invested because they turned 21 in an IPO year andotherwise would not have invested. Robust standard errors, clustered at most recent county of represen-tation level. *10%, **5%, ***1%. Columns (1) and (3) are Long Parliament MPs only. Columns (2) and (4)pool all Long Parliament and 1628 Parliament MPs. Columns (1) and (2) are the full sample; columns (3)and (4) restrict only to MPs representing the core lay subsidy constituencies.


Dow




ber 2019

subsequently rebel (Panel B). This yields a Wald IV estimate ofthe effect of shareholding in overseas joint stock on the proba-bility of rebellion of those that invested because they came of agein an overseas IPO year of more than 95.4 percentage points(Panel C).32

Table VIII, Panel A compares MPs who came of age during anIPO year only to a sample of other MPs that came of age withinfive years of an IPO year, including indicator variables forwhether an MP came of age each of a minimum of one to fiveyears before or after an IPO year. The comparison is local inthe sense that the sample is restricted to MPs who came of agewithin a five-year window on either side, with the excluded cat-egory therefore being MPs who came of age five years after anIPO. Figure IV plots the OLS coefficients on the indicator vari-ables of columns (1) and (5). Notice that there is a spike in theprobability of investment in the IPO year: MPs who come of age inan IPO year were 25.5 percentage points more likely to invest inthis local comparison. There is also some increase in propensity toinvest for those who came of age a year before (-1) or two yearsbefore (-2) as well (not surprising given that these cohorts werealso old enough to invest), but MPs who come of age in subsequentyears are not significantly more likely to do so. This translatesinto an increased probability of rebellion of around 88.9 percent-age points among those who invested in shares because they cameof age in an IPO year (column (9)). Table VIII, Panel A shows thatthe reduced-form relationships are robust to adding individualand wealth controls (columns (2), (4), (6), (8)), and pooling MPsfrom the 1628 Parliament (columns (3)–(4),(7)–(8)). This yields IVestimates of the effect of shareholding on the propensity to rebelamong those who invested because they came of age in an IPOyear that range between 55.7 to 99.1 percentage points (columns(9)–(12)). F-tests of the instrument in the first-stage regressionsare sufficient to avoid weak instrument pathologies across theselocal specifications (columns (1)–(4)).

Panels B and C further subset the data by mercantile en-dowment. Consistent with the presence of an alignment effect of

32. In the Wald comparison in the general sample, the relatively low F-statisticson the instrument raises the possibility of biases due to weak instruments. I there-fore report the reduced form (Panel B), which is unbiased, and LIML coefficientsand standard errors, which are less subject to weak instrument pathologies. Thebias, if it exists, should lead LIML to be biased toward OLS. The 2SLS estimates aresimilar but more precise.


Dow




ber 2019

TA

BL

EV

III

FIR

ST-S

TA

GE,

RE

DU

CE

D-F

OR

M,

AN

DL

OC

AL

IVE

ST

IMA

TE

SO

FE

FF

EC

TO

FS

HA

RE

HO

LD

ING

ON

DE

CIS

ION

TO

RE

BE

L

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

(12)

OL

SO

LS

OL

SO

LS

OL

SO

LS

OL

SO

LS

IV-L

IML

IV-L

IML

IV-L

IML

IV-L

IML

Ou

tcom

eO

ver

seas

inves

tor

Over

seas

inves

tor

Over

seas

inves

tor

Over

seas

inves

tor

Reb

el,

Civ

ilW

ar

Reb

el,

Civ

ilW

ar

Reb

el,

Civ

ilW

ar

Reb

el,

Civ

ilW

ar

Reb

el,

Civ

ilW

ar

Reb

el,

Civ

ilW

ar

Reb

el,

Civ

ilW

ar

Reb

el,

Civ

ilW

ar

Pan

elA

:A

llM

Ps

Inves

tor

inov

erse

as

join

tst

ock

0.8

89**

0.5

57

0.9

91**

0.7

12*

[0.3

69]

[0.3

78]

[0.4

04]

[0.3

87]

Att

ain

edad

ult

hoo

din

IPO

yea

r0.2

55**

*0.2

35**

*0.2

35**

*0.2

23**

*0.2

23**

*0.1

23

0.2

33**

*0.1

59*

[0.0

68]

[0.0

71]

[0.0

63]

[0.0

65]

[0.0

81]

[0.0

94]

[0.0

81]

[0.0

88]

F-t

est

inst

rum

ent

14.2

11.0

913.7

811.6

9P

rob>

F0.0

00

0.0

00

0.0

00

0.0

00

R2

0.0

30.1

00.0

20.0

90.0

20.1

10.0

20.1

0O

bse

rvati

ons

482

479

518

514

482

479

518

514

481

478

518

514

Pan

elB

:N

on-m

erca

nti

leM

Ps

Inves

tor

inov

erse

as

join

tst

ock

1.3

87*

0.8

57

1.7

15*

1.1

60*

[0.8

30]

[0.6

06]

[0.9

96]

[0.6

70]

Att

ain

edad

ult

hoo

din

IPO

yea

r0.1

44**

0.1

52**

0.1

26**

0.1

40**

0.1

95**

0.1

20

0.2

16**

0.1

62*

[0.0

65]

[0.0

72]

[0.0

61]

[0.0

65]

[0.0

94]

[0.0

94]

[0.0

94]

[0.0

91]

R2

0.0

20.0

60.0

20.0

50.0

20.1

00.0

30.1

0O

bse

rvati

ons

397

395

428

425

397

395

428

425

396

394

428

425

Pan

elC

:M

Ps

ap

pre

nti

ced

mer

chan

tor

fath

erm

erch

an

tIn

ves

tor

inov

erse

as

join

tst

ock

0.1

61

0.1

08

0.1

69

0.1

75

[0.5

41]

[0.5

54]

[0.5

25]

[0.5

85]

Att

ain

edad

ult

hoo

din

IPO

yea

r0.5

02**

0.5

48*

0.5

15**

0.5

13

0.0

81

0.0

59

0.0

87

0.0

90

[0.2

08]

[0.3

20]

[0.2

05]

[0.3

09]

[0.2

85]

[0.3

51]

[0.2

84]

[0.3

37]

R2

0.1

70.2

90.1

80.2

70.0

60.1

80.0

60.1

9O

bse

rvati

ons

85

84

90

89

85

84

90

89

85

84

90

89


Dow




ber 2019

TA

BL

EV

III

(CO

NT

INU

ED)

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

(12)

OL

SO

LS

OL

SO

LS

OL

SO

LS

OL

SO

LS

IV-L

IML

IV-L

IML

IV-L

IML

IV-L

IML

Ou

tcom

eO

ver

seas

inves

tor

Over

seas

inves

tor

Over

seas

inves

tor

Over

seas

inves

tor

Reb

el,

Civ

ilW

ar

Reb

el,

Civ

ilW

ar

Reb

el,

Civ

ilW

ar

Reb

el,

Civ

ilW

ar

Reb

el,

Civ

ilW

ar

Reb

el,

Civ

ilW

ar

Reb

el,

Civ

ilW

ar

Reb

el,

Civ

ilW

ar

Wea

lth

an

dot

her

ind

ivid

ual

con

trol

sN

oY

esN

oY

esN

oY

esN

oY

esN

oY

esN

oY

es

Sam

ple

:yea

rof

ad

ult

hoo

dw

/in

5yea

rsof

IPO

sin

:L

ong

Parl

Lon

gP

arl

LP

+1628

LP

+1628

Lon

gP

arl

Lon

gP

arl

LP

+1628

LP

+1628

Lon

gP

arl

Lon

gP

arl

LP

+1628

LP

+1628

Not

es:

Col

um

ns

(1)–

(4)

pro

vid

ees

tim

ate

sof

the

firs

tst

age

effe

ctof

turn

ing

21

inan

IPO

yea

ron

an

MP

’sd

ecis

ion

toin

ves

tin

over

seas

share

s.C

olu

mn

s(5

)–(8

)p

rovid

eth

ere

du

ced

-for

mef

fect

oftu

rnin

g21

inan

IPO

yea

ron

aM

P’s

dec

isio

nto

rebel

inth

eC

ivil

War.

Col

um

ns

(9)–

(12)

pro

vid

esth

eIV

-LIM

Les

tim

ate

.T

he

sam

ple

isre

stri

cted

toth

ose

MP

sth

at

turn

ed21

wit

hin

5yea

rsof

an

over

seas

IPO

,an

dall

regre

ssio

ns

incl

ud

ed

um

my

vari

able

sfo

rco

min

gof

age

for

each

of1–5

yea

rsbef

ore

an

d1–4

yea

rsaft

er.

Th

eref

ore

the

excl

ud

edca

tegor

yis

five

yea

rsaft

era

IPO

.R

obu

stst

an

dard

erro

rscl

ust

ered

at

the

cou

nty

ofm

ost

rece

nt

con

stit

uen

cy.

*10%

,**

5%

,**

*1%

.In

div

idu

al

con

trol

sin

clu

de

fath

erJS

inves

tor,

ap

pre

nti

ced

orfa

ther

mer

chan

t,gen

try

bef

ore

Tu

dor

s,in

her

ited

cou

rtti

es,

log.

min

imu

md

ista

nce

offa

ther

’sad

dre

ssto

Lon

don

,n

um

ber

offa

ther

’sad

dre

sses

,att

end

eda

Pu

rita

nS

emin

ary

,an

dw

ealt

hen

dow

men

tsth

at

incl

ud

edin

her

ited

am

an

or,

inh

erit

edla

nd

,h

eir,

fath

erk

nig

ht

orbaro

net

,fa

ther

nob

le,

an

dex

per

ien

ced

ward

ship

.‘‘L

ong

Parl

’’im

pli

esth

eL

ong

Parl

iam

ent

sam

ple

ofM

Ps

only

,w

hil

e‘‘L

P+

1628’’

poo

lsM

Ps

from

the

1628

Parl

iam

ent.


Dow




ber 2019

shares, the results suggest that though both MPs with mercan-tile and nonmercantile endowments invest more if they come ofage in an IPO year (columns (1)–(4)), the effect of shareholding isto increase the propensity to rebel among nonmerchants only.This is true in both the reduced form (columns (5)–(8)) and theIV estimates (columns (9)–(12)). Figure V once again plots theOLS coefficients of indicators of the year an MP came of agerelative to an IPO year, this time subsetting by mercantile en-dowment. Notice that while both groups were more likely toinvest in an IPO year than in other years (top panel), a spikein the propensity to rebel among those who come of age in anIPO year is only discernible for nonmerchants (bottom panel).The combination of an increase in the propensity to invest

FIGURE IV

Overseas Share Investment and Rebellion by Age of Adulthood before and afteran IPO year: Long Parliament MPs

OLS coefficients of probability of investment and rebellion on indicator var-iables for whether a Long Parliament MP turned 21 in the year of an overseasIPO (0) or the least number of years came of age before an overseas IPO (�1 to�5) and after an overseas IPO (+1 - +4). The comparison is restricted only toMPs that fall within the five-year-before and -after window, and thus the ex-cluded category is those who came of age five years after an IPO. Standarderrors clustered at the county level. Notice that MPs that come of age in an IPOyear (0) are more likely to invest and to rebel than those that come of age after(+1). Those that come of age before an IPO year (e.g., �2, �1) also could legallyinvest but did so at lower rates.


Dow




ber 2019

among merchants who came of age in an IPO year, coupledwith a lack of effect on their political allegiance, is inconsis-tent with the possibility that all those that came of age in anIPO year were antimonarchical in general, but remains con-sistent with the alignment effect of shares specifically onnonmerchants.

One may be concerned that adulthood in an IPO yearmight happen to coincide with other characteristics that alsomight influence shareholding. Online Appendix Figure A1.8shows that in the same local reduced-form comparison, andconsistent with the covariate balance in means in Table VI,there are no spikes in a range of placebo endowments forMPs that come of age in an IPO year, including whether the

FIGURE V

Overseas Share Investment and Rebellion before and after an IPO year, byMercantile Endowment

OLS coefficients of probability of investment and rebellion on indicatorvariables for whether a Long Parliament MP turned 21 in the year of an over-seas IPO (0) or the least number of years came of age before an overseas IPO(�1 to �5) and after an overseas IPO (+1 - +4), separating MPs who wereapprenticed merchants or whose fathers were merchants (L) and others (R).The comparison is restricted only to MPs that fall within the five-year-beforeand -after window, and thus the excluded category is those who came of age fiveyears after an IPO. Standard errors clustered at the county level. Notice thatwhile merchants and nonmerchants were more likely to invest if they came ofage in an IPO year, only nonmerchants are more likely to rebel.


Dow




ber 2019


MP’s father invested in joint stock and a range of wealthendowments.33

Online Appendix Figure A1.9 further examines whetherthere are spikes among MPs who come of age in an IPO year inother channels (top row) and outcomes (bottom row). The figureshows that MPs that come of age in IPO years are not any morelikely to invest in domestic joint stock (an indicator of risk pref-erence), to hold or avoid holding royal office (an indicator of preex-isting ideology), or to have higher income at the time of the CivilWar. However, consistent with the OLS matching results, theredo appear to be spikes in measures of political allegiance justprior and just following the Civil War, particularly the probabilityof subscribing to the loan of 1642 in defense of London and ofserving in the Rump Parliament of victors.

As a further robustness check, instead of looking only at MPswho came of age within a five-year window around IPO years,Online Appendix Table A1.9 presents a regression of LongParliament MPs who came of age at any time, with the full setof individual and constituency controls and county fixed effects,and adding global quadratic and quartic polynomial controls forthe MP’s age. To increase precision, I include as a second instru-ment the interaction of whether an MP came of age in an IPOyear and the minimum distance of the MP’s father’s residence toLondon. As columns (1)–(4) suggest, individuals who turned 21 inan IPO year are even more likely to invest if their fathers hadlived closer to London. However, the proximity of an MP’s father’sresidence to London has no separate direct effect beyond this. Theeffect of shareholding on rebellion is also robust to including qua-dratic and quartic controls for the age of the MP, which also ap-pears to have no robust direct relationship on the propensity toinvest or rebel (columns (1)–(8)). These comparisons yield IV-LIML estimates of the effects of shareholding on the propensityto rebel among those who invested because they came of age inIPO years of between 58.7 and 66.9 percentage points, which arereassuringly in line with the estimates based on the localcomparisons.

33. The one possible exception is that MPs who come of age in IPO years are lesslikely to have fathers who were nobles. As discussed, Online Appendix Table A1.11uses corroborative data from the House of Lords to show that nobles also appearmore likely to invest and rebel in IPO years.


Dow




ber 2019




V.C. Assessing Economic Significance

I can use the conventional, lower bound, and LATE estimatesof the effect of shares to perform a counterfactual exercise. Asdiscussed, MPs were faced with the stark choice of supportingthe Crown or Parliament during the Civil War. Since thischoice is symmetric—not making a choice is not an option—those pushed over the 50 percent probability threshold of sup-porting Parliament due to holding shares can be thought of aslikely switchers of allegiance from the Crown.

With 56 percent of Long Parliament MPs choosing to rebel inthe Civil War, and thus 44 percent who chose to be royalists(Table I), the margin of the majority favoring parliamentary su-premacy was a slim 31 members. Figure VI shows the distribu-tion of predicted probabilities of rebellion among MPs, as well as acounterfactual distribution had shares not shifted the allegianceof shareholders. As the figures suggest, across specifications,

FIGURE VI

Probability Distributions of Rebellion with and without Share Effects onShareholders

The bold probability distribution provides the predicted probability of re-bellion for each specification. The dashed probability distribution subtracts theestimated effect of shareholding on shareholders to generate a counterfactualdistribution. Beginning with the top left, each figure sequentially adds controls.Observe that, across specifications, shareholding appears to shift moderates infavor of rebellion.


Dow




ber 2019

overseas share investment shifted the affiliations of MPs whowere moderates, ex ante, rather than inframarginal extremists.I can further exploit the fact that individuals had to choose be-tween king and Parliament and use the 50 percent threshold onthe probability of rebellion to examine who these switchers werelikely to be. Online Appendix Tables A1.12 and A1.13 exploit boththe OLS and IV estimates to generate a list of compliant switch-ers: shareholder MPs who ultimately supported Parliament but,all else equal, were actually more likely to have supported theCrown in the absence of shares. The MPs are ordered by alle-giance, with those predicted most likely to have been royalistsin the absence of shares at the bottom. In the basic specificationwith personal endowment controls, 39 MPs who rebelled weremore likely to have been royalist in the absence of shares, aneffect sufficient to push support for Parliamentary supremacyfrom a minority position to the majority (Online AppendixTable A1.12). As Table A1.12 and the expanded set of potentialswitchers in Table A1.13 suggest, the full set of controls in the laysubsidy subsample yields 14 compliant switchers, whereas thelocal IV and Wald estimates yield 10 and 20, respectively. Thus,shareholding appears to have consolidated majority support forParliament among moderate MPs and may in fact have been piv-otal in its creation.

The majority of these compliant switcher MPs came fromgentry, rather than mercantile, backgrounds. A number also as-sumed important court offices prior to the Civil War. Eight of the39 switchers were knights of the shire—the holders of the countyseat—that tended to have the largest franchises and the mostprestige among rural landowners (see also Online AppendixTable A2.3). While a few were involved in the profitable new trad-ing companies, like the East India Company, a large majoritywere invested in unprofitable colonization projects, with sharesbringing acres of land in the New World, but whose charters andinvestments were subject to Crown expropriation. A remarkablefeature of Online Appendix Table A1.12 is that despite looking atmembers who otherwise would have supported the Crown in theabsence of shares, the table contains all of the famous FiveMembers that were identified by the king to be ringleaders ofParliament’s legislative challenges to his prerogative rights.The king’s illegal armed entry into Parliament to arrest theFive for treason in January 1642, which later led to the mobili-zation of the London militia in defense of Parliament, is often


Dow




ber 2019










seen as a major step toward the outbreak of violent conflict(Hexter 1941). Three of the Five Members—Denzil Holles, JohnPym, and William Strode—are among those predicted most likelyto be royalists in the absence of shares. The table also includesfour of the six members of the subcommittee that framed theGrand Remonstrance, which laid down the reforms thatParliament demanded for the king. Of those on both lists, JohnPym in particular is seen by historians as a major figure in ini-tially organizing the coalition in favor of Parliamentary control(Hexter 1941; Hill 1961).34 Yet these results suggest these leadersof the opposition might have actually favored support for theCrown in the absence of shares. Instead of being radical extrem-ists, it may be that the leaders of the push for constitutionalreform were instead effective at creating a coalition precisely be-cause their interests were aligned, through both their endow-ments and their share investments, with both landowners andoverseas traders.35

VI. Discussion

The Civil War was a tragedy for England, with an estimated100,000 deaths in battle and a further 100,000 civilian deaths(out of England’s population of five million) attributed to war

34. Recent historical work downplays the role of the leadership in the Commonsin the Long Parliament, pointing to social and client ties between these individualsand the nobility in the House of Lords (e.g., Adamson 2007). Though insightful,these studies do not address the issue that just as court ties were seen as effectivemeans to secure property rights, such client ties with the nobility were also endog-enous and may have been strategically developed. In the House of Lords during theCivil War, Firth (1910, p. 60) describes how the ‘‘future leaders of the popular partyin the two Houses of the Long Parliament were brought together and learned to co-operate’’ in overseas joint stock companies. As Online Appendix Table A1.11 shows,members of the House of Lords also invested more in IPO years and appear morelikely to rebel as a result.

35. Online Appendix Table A1.14 shows those shareholder MPs who werepushed above 50 percent probability of rebellion as a result of shares but did notultimately rebel (based upon the probit regression with personal endowment con-trols in Table III (column (1)). As the table suggests, of these 16 royalists, 11 heldcourt offices or customs farms. These MPs appear to have adopted the parallelstrategy of cultivating court ties to secure their property rights. However, the listalso includes some of the surprise turncoats of the Civil War, like Robert, Lord Rich,an investor in the Providence Island Company, whose father, the Earl of Warwick,was an important leader of the rebel faction in the House of Lords but who himselfdefected to the royalists.


Dow




ber 2019



and war-related famine and disease (Clodfelter 2008). YetEngland since then has not experienced a major violent conflictaimed at seizing political control of the state.36 This is in sharpcontrast even to other contemporary consolidated WesternEuropean nation-states, such as France, that experienced masscivil upheavals in the 1790s and Spain, which experienced suchconflict as late as the 1930s, in both cases ending up with militarydictatorships. Furthermore, though strong disagreements ex-isted between the Parliamentary victors of the Civil War, therewould be little fratricidal bloodletting between rival revolution-ary groups, a sadly common aspect of revolutions that followed.

Not surprisingly, then, the question of how a novel, broad,and ultimately stable coalition in favor of parliamentary suprem-acy in government emerged in England has proved to be amongthe most enduring in political economy and institutional econom-ics. What do we learn from England’s experience? Nuancing anumber of important political economy interpretations ofEngland’s revolution, that have hypothesized that changes tothe existing distribution of wealth drove political change by cre-ating a newly enriched middle class (Moore 1966; Rajan andZingales 2003; Acemoglu, Johnson and Robinson 2005), it insteadappears that the introduction of shares aligned incentives by pro-viding a means through which potentially anyone could avail offuture opportunities. Reformers not only possessed a majority inParliament, but as I have shown, even when individuals wereforced to pick a side, reformers enjoyed a broad coalition ofwealthholders that spanned initial social distinctions betweenmerchants and nonmerchants, devout and worldly. The forma-tion of such an encompassing coalition appears to have raisedreformers’ expectations that future majorities would choose pol-icies more akin to their own than a single, potentially capriciousindividual, and explains why they pushed for an institutionalcommitment device—the establishment of parliamentary control

36. The 1688 Revolution was later called ‘‘Glorious’’ precisely because it wasvirtually uncontested and bloodless in England. Similarly other political uprisingsaimed at constitutional change in 1685, 1714, and 1745 failed to attract supportwithin England. Monmouth’s rebellion in 1685 led to1,384 rebel dead in battle, with333 executed in the Bloody Assizes, and 400 royalist losses (at most). The Jacobiterebellion in 1715 mobilized 7,000 Scots and at most 3,000 English Jacobites but wasdriven out of Preston by a force of 2000. The ‘‘45’’ mobilized 8,000 Scots at peak(Clodfelter 2008).


Dow




ber 2019

of government—rather than fighting for narrower sectarian orclass objectives.37

Because they depended on their strength on the alignment ofinterests of a broad coalition of wealthholders, it is perhaps notsurprising that the parliamentary victors of the Civil War did notradically alter domestic property rights (see Online AppendixFigure A1.3). However, England’s overseas policies did changedramatically. Consistent with the importance of shared exposureto overseas constitutional risk in shaping the coalition, England’snew rulers invested heavily in a particular set of public goods—the Royal Navy—with the number of naval warships rising from56 at the end of the English Civil War in 1648 to 133 within sevenyears (Rodger 2004). These investments laid the basis for Britishnaval supremacy (Online Appendix Figure A1.2) and accompa-nied a series of mercantilist wars fought to protect overseas ex-pansion and redirect commerce from rivals. Naval supremacyhelped ensure London’s emergence as the most important tradinghub of Europe by the early eighteenth century and Britain’sacquisition of an empire that spread the influence of its institu-tions—including the common law, parliamentary paramountcy ingovernment, and the Bill of Rights—around the world (Ferguson2002; Ormrod 2003; de la Escosura 2004). Shareholders holdingassets overseas benefited from a dramatic rise in the relativevalue of land in the New World after the Civil War, even whilethe return on land in England remained relatively unchanged(Online Appendix Figure A1.3).

37. Some see England as entering a military dictatorship when the RumpParliament was dismissed by Oliver Cromwell in 1653. In fact, this dismissal oc-curred because the Rump was seeking to establish itself in perpetuity without al-lowing fresh elections. Instead, according to Schama (2001, p. 232), ‘‘the[constitution of Cromwell’s Protectorate] corresponded to the proposals set out bythe most advanced parliamentarians of the 1640s, and for that matter to whatwould actually come to pass after the next round of revolution in 1688–90’’ (seealso Woolrych 2003). The Cavalier Parliament ceded its right to call and dismissitself back to the king shortly after the restoration of the monarchy in 1660, butretained enhanced control over state finances that dramatically limited the king’sability to act independently (see also Murrell (2009)). Indeed, unlike prior to theCivil War, Parliament met in virtually every year following the Restoration (OnlineAppendix Figure A1.1). Charles II was able to reassert limited monarchical power,but only by adopting policies he did not personally favor but had the support of theHigh Tories who had replaced the Whigs as a new majority in Parliament (Harris2005). In contrast, James II’s attempts to impose policies that lacked support evenfrom High Tories precipitated the Glorious Revolution in 1688 and consolidatedparliamentary supremacy (Harris 2006; Pincus and Robinson 2011).


Dow




ber 2019







England’s post–Civil War government remained that of thewealthy, but not a stable subsection of the wealthy, nor was itclosed to entrants. The first political parties (Whig and Tory)transcended traditional cleavages (town versus country, landedversus merchant) and instead coalesced around investments inemergent joint stock companies (Carruthers 1999). It is possiblethat the development of active secondary asset markets that oc-curred between the Civil War and Glorious Revolution may haveallowed both the winners and losers from the Civil War to reallo-cate their portfolios in favor of those investments benefiting froman assertive foreign policy, allowing a further broadening of thecoalition in favor of rule by the majority of wealthholders inParliament, rather than by the king (Carlos and Jha 2015).Though debate was heated, particularly over religion and theburdens of state finance (Stasavage 2003), it was much less vio-lently conflictual than before. Rather than becoming an oligarchyof overseas investors, the introduction of financial assetholdingand the subsequent development of secondary markets may havehelped align disparate interests that led post–Civil War Englandon a trajectory toward ministerial responsibility (Cox 2012), do-mestic peace, and broadly representative government.

More broadly, the failure to align the incentives of self-interested groups in favor of beneficial reform is often considereda major cause of civil conflict and persistent underdevelopmentaround the world. In England, the introduction of shares allowedindividuals possessing mercantile human capital to sell thestream of benefits and the risks of their human capital to nonmer-chants, even though human capital was (and remains) conven-tionally seen as nontradable. Furthermore, this exposure appearsto have had important benefits in mitigating class conflict andfostering broader-based political reform. The English examplesuggests the possibility for a political multiplier of financialmarket development on economic growth.

Indeed, like in revolutionary England, financial revolutionspreceded economic development in two other revolutionarystates—the United States and Japan—that subsequently ledthe world in GDP growth (Sylla 2002). In fact, partly inspiredby the English example, reformers in first the postrevolutionaryUnited States and later Meiji Japan introduced banking systemsthat similarly provided share ownership and common exposure topolitical risk to groups that might have otherwise resisted furtherreforms (Jha 2012). Potential losers likely to violently resist


Dow




ber 2019

modernization, like members of the samurai caste, insteadbecame the shareholders of banks and thus credible beneficiariesof reduced political risk and future growth (Jha 2012; Jha,Mitchener, and Takashima 2015). Indeed, the introduction ofnew financial assets may be successful in mitigating contempo-rary ethnic conflict (Jha and Shayo 2015). The question of how abroad coalition in favor of parliamentary supremacy emerged inrevolutionary England is among the oldest and most famous inpolitical economy. There may yet be new lessons it can teach us.

Stanford Graduate School of Business

Supplementary Material

An Online Appendix for this article can be found at QJEonline (qje.oxfordjournal.org).

References

Acemoglu, Daron, Simon Johnson, and James Robinson, ‘‘The Rise of Europe:Atlantic Trade, Institutional Change and Economic Growth,’’ AmericanEconomic Review, 95 (2005), 546–579.

Acemoglu, Daron, and James Robinson, Why Nations Fail (New York: CrownBusiness, 2012).

Adamson, John, The Noble Revolt: The Overthrow of Charles I (London:Weidenfeld and Nicholson, 2007).

Andrews, Kenneth R., Drake’s Voyages: A Reassessment of Their Place inElizabethan Maritime Expansion (London: Weidenfeld and Nicolson, 1967).

Barber, Brad M., and Terrance Odean, ‘‘All that Glitters: The Effect of Attentionand News on the Buying Behavior of Individual and Institutional Investors,’’Review of Financial Studies, 21 (2006), 785–818.

Benmelech, Efraim, and Tobias J. Moskowitz, ‘‘The Political Economy of FinancialRegulation: Evidence from US State Usury Laws in the 19th Century,’’Journal of Finance, 65 (2010), 1029–1073.

Blackstone, Sir William, Commentaries on the Laws of England, vol. 1 (Oxford:Clarendon Press, 1769).

Brenner, Robert, Merchants and Revolution: Commercial Change, PoliticalConflict and London’s Overseas Traders, 1550–1653 (Princeton, NJ:Princeton University Press, 1993).

Brunton, Douglas, and David H. Pennington, Members of the Long Parliament(London: George Allen and Unwin, 1954).

Burgess, Glenn, The Politics of the Ancient Constitution: An Introduction toEnglish Political Thought, 1603–1642 (University Park: PennsylvaniaState University Press, 1992).

Carlos, Ann M., and Saumitra Jha, ‘‘Stockmarkets and Political Conflict:Evidence from Revolutionary England,’’ Working Paper, 2015.

Carlos, Ann M., Jennifer Key, and Jill L. Dupree, ‘‘Learning and the Creation ofStock Market Institutions: Evidence from the Royal African and Hudson’sBay Companies, 1670–1700,’’ Journal of Economic History, 58 (1998), 318–344.

Carlyle, Thomas, Oliver Cromwell’s Letters and Speeches, vol. 3 (London:Chapman and Hall, 1897 [1845]).


Dow




ber 2019


Carruthers, Bruce, City of Capital: Politics and Markets in the English FinancialRevolution (Princeton, NJ: Princeton University Press, 1999).

Charles I and John Gauden, Eikon Basilike: The Portraiture of His SacredMajesty in His Solitude and Sufferings (Ithaca, NY: Cornell UniversityPress, 1966 [1649]).

Chaudhuri, K. N., The English East India Company: The Study of an Early Joint-Stock Company 1600–1640 (New York: Augustus Kelley, 1965).

Clark, Gregory, ‘‘The Political Foundations of Modern Economic Growth:England, 1540–1800,’’ Journal of Interdisciplinary History, 26 (1996), 563–588.

Clodfeiter, Michael, War and Armed Conflicts: A Statistical Encyclopedia ofCasualty and Other Figures, 1494–2007 (Jefferson, NC: McFarland, 2008).

Coval, Joshua, and Tobias Moskowitz, ‘‘Home Bias at Home: Local EquityPreference in Domestic Portfolios,’’ Journal of Finance, 65 (1999), 2045–2073.

Cox, Gary W., ‘‘Was the Glorious Revolution a Constitutional Watershed?,’’ Journalof Economic History, 72 (2012), 567–600.

de la Escosura, Leandro Prados, ed., Exceptionalism and Industrialisation:Britain and its European Rivals: 1688–1815 (Cambridge: CambridgeUniversity Press, 2004).

de Lara, Yadira Gonzalez, Avner Greif, and Saumitra Jha, ‘‘The AdministrativeFoundations of Self-Enforcing Constitutions,’’ American Economic Review,98 (2008), 105–109.

Fearon, James D., ‘‘Bargaining over Objects that Influence Future BargainingPower,’’ Mimeo, APSA general meetings, 1996.

Ferguson, Niall, Empire: The Rise and Decline of the British World Order and theLessons for Global Power (London: Basic Books, 2002).

Firth, Charles Harding, The House Of Lords During the Civil War (London:Longmans, Green, 1910).

Firth, Charles Harding, and R. S. Rait, eds., Acts and Ordinances of theInterregnum, 1642–1660, 1911, available at http://www.british-history.ac.uk/report.aspx?compid=55730.

Gardiner, Samuel Rawson, ed. , Parliamentary Debates in 1610, Edited, from theNotes of a Member of the House of Commons, vol. 81 (London: CamdenSociety, 1862).

Gras, N. S. B., ‘‘Tudor ‘Book of Rates’: A Chapter in the History of the EnglishCustoms,’’ Quarterly Journal of Economics, 26 (1912), 766–775.

Grassby, Richard, ‘‘The Personal Wealth of the Business Community inSeventeenth-Century England,’’ Economic History Review, 23 (1970),220–234.

Greif, Avner, ‘‘Commitment, Coercion and Markets: The Nature and Dynamics ofInstitutions Supporting Exchange,’’ in Handbook for New InstitutionalEconomics, Claude Menard and Mary M. Shirley, eds. (Norwell, MA:Kluwer Academic, 2005), 727–788.

Guiso, Luigi, and Tullio Jappelli, ‘‘Awareness and Stockmarket Participation,’’ Reviewof Finance, 9 (2005), 537–567.

Haber, Stephen, and Enrico Perotti, ‘‘The Political Economy of FinancialSystems,’’ Technical Report, Tinbergen Institute Discussion Paper, 2008.

Hakluyt, Richard, Voyages, vol. 1 (London: Dent, 1962 [1589]).Harris, Ron, Industrializing English Law: Entrepreneurship and Business

Organization, 1720–1844 (Cambridge: Cambridge University Press, 2000).———, ‘‘Law, Finance and the First Corporations,’’ in Global Perspectives on the

Rule of Law, James J. Heckman, Robert L. Nelson, and Lee Cabatingan, eds.(London: Routledge, 2009).

Harris, Tim, Restoration: Charles II and His Kingdoms (London: Penguin, 2005).———, Revolution: The Great Crisis of the British Monarchy, 1685–1720 (London:

Penguin, 2006).Hexter, J. H., The Reign of King Pym (Cambridge, MA: Harvard University Press,

1941).Hill, Christopher, The Century of Revolution: 1603–1714 (New York: Routledge,

1961).Imbens, Guido W., ‘‘Sensitivity to Exogeneity Assumptions in Program

Evaluation,’’ American Economic Review, 93 (2003), 126–132.


Dow




ber 2019

http://www.british-history.ac.uk/report.aspx?compid=55730

http://www.british-history.ac.uk/report.aspx?compid=55730

Jha, Saumitra, ‘‘Sharing the Future: Financial Innovations and Innovators inSolving the Political Economy Challenges of Development,’’ in Institutionsand Comparative Economic Development, Masahiko Aoki, Timur Kuran, andGerard Roland, eds. (Basingstoke: Palgrave Macmillan, 2012).

Jha, Saumitra, Kris Mitchener, and Masanori Takashima, ‘‘Swords into BankShares: Finance, Conflict and Political Reform in Meiji Japan,’’ StanfordGSB working paper, 2015.

Jha, Saumitra, and Moses Shayo, ‘‘Voting for Peace in a Conflict Zone: The Effectsof Exposure to Financial Markets,’’ Stanford GSB working paper, 2015.

Kaustia, Markku, Samuli Knupfer, and Sami Torstila, ‘‘Stock Ownership andPolitical Behavior: Evidence from Demutualizations,’’ Management Science(2015), forthcoming.

Keeler, Mary Frear, The Long Parliament: 1640–1641, A Biographical Study of ItsMembers (Philadelphia, PA: American Philosophical Society, 1954).

Kelsey, Harry, Sir Francis Drake: The Queen’s Pirate (New Haven, CT: YaleUniversity Press, 1998).

Lindert, Peter H., and Jeffrey G. Williamson, ‘‘Revising England’s Social Tables1688–1812,’’ Explorations in Economic History, 19 (1982), 345–408.

McGrath, Patrick, Papists and Puritans under Elizabeth I (New York: Walker,1967).

Merton, Robert, ‘‘A Simple Model of Capital Market Equilibrium with IncompleteInformation,’’ Journal of Finance, 42 (1987), 483–510.

Mitchell, William M., The Rise of the Revolutionary Party in the English House ofCommons (New York: Columbia University Press, 1957).

Moore, Barrington, Social Origins of Dictatorship and Democracy: Lordand Peasant in the Making of the Modern World (Boston: Beacon, 1993[1966]).

Morrill, John S., The Nature of the English Revolution (New York: Longman,1993).

Murrell, Peter, ‘‘Design and Evolution in Institutional Development: TheInsignificance of the English Bill of Rights,’’ Mimeo, Maryland, 2009.

North, Douglass C., and Barry R. Weingast, ‘‘Constitutions and Commitment:Evolution of Institutions Governing Public Choice,’’ Journal of EconomicHistory, 49 (1989), 803–832.

Ormrod, David, The Rise of Commercial Empires: England and the Netherlandsin the Age of Mercantilism, 1650–1770 (Cambridge: Cambridge UniversityPress, 2003).

Oster, Emily, ‘‘Unobservable Selection and Coefficient Stability: Theory andEvidence,’’ NBER Working Paper, Chicago Booth, 2014.

Pincus, Steven C. A., 1688: The First Modern Revolution (New Haven, CT: YaleUniversity Press, 2009).

Pincus, Steven C. A., and James A. Robinson, ‘‘What Really Happened in theGlorious Revolution?,’’ NBER Working Paper 17206, 2011.

Plumb, J. H., ‘‘The Growth of the Electorate in England from 1600 to 1715,’’ Pastand Present, 45 (1969), 90–116.

Quinn, Stephen, ‘‘The Glorious Revolution’s Effect on British Private Finance: AMicrohistory,’’ Journal of Economic History, 61 (2001), 593–615.

Rabb, Theodore K., Enterprise and Empire: Merchant and Gentry Investment inthe Expansion of England, 1575–1630 (Cambridge, MA: Harvard UniversityPress, 1967).

———, Jacobean Gentleman: Sir Edwin Sandys, 1561–1629 (Princeton, NJ:Princeton University Press, 1998).

Rajan, Raghuram G., ‘‘Competitive Rent Preservation, Reform Paralysis, and thePersistence of Underdevelopment,’’ American Economic Journal: Macroeconomics,1 (2009), 178–218.

Rajan, Raghuram G., and Luigi Zingales, Saving Capitalism from the Capitalists:Unleashing the Power of Financial Markets to Create Wealth and SpreadOpportunity (New York: Crown Business, 2003).

Rodger, N. A. M., The Command of the Ocean: A Naval History of Britain, 1649–1815, vol. 2 (London: Penguin, 2004).

Schama, Simon, A History of Britain: The Wars of the British, 1603–1776 (NewYork: Hyperion, 2001).


Dow




ber 2019

Scott, William Robert, The Constitution and Finance of English, Scottish andIrish Joint-Stock Companies to 1720: Companies for Foreign Trade,Colonization, Fishing and Mining, vol. 2 (Cambridge: CambridgeUniversity Press, 1910).

Smith, David L., The Stuart Parliaments, 1603–1689 (London: Arnold, 1999).Stasavage, David, Public Debt and the Birth of the Democratic State: France and

Great Britain, 1688–1789 (Cambridge: Cambridge University Press, 2003).Stone, Lawrence, ‘‘The Bourgeois Revolution of Seventeenth Century England

Revisited,’’ Past and Present, 109 (1985), 44–54.Sussman, Nathan, and Yishay Yafeh, ‘‘Constitutions and Commitment: Evidence

on the Relation between Institutions and the Cost of Capital,’’ TechnicalReport, CEPR Discussion Papers, 2004.

Sylla, Richard, ‘‘Financial Systems and Economic Modernization,’’ Journal ofEconomic History, 62 (2002), 277–291.

Tawney, R. H., ‘‘The Rise of the Gentry, 1558–1640,’’ Economic History Review, 11(1941), 1–38.

Whitelocke, Bulstrode, Memorials of the English Affairs (London: EighteenthCentury Collections Online: Range 9539, 1732 [1682]).

Woolrych, Austin, ‘‘The Cromwellian Protectorate: A Military Dictatorship?,’’ inCromwell and the Interregnum: The Essential Readings, David L. Smith , ed.(Malden, MA: Blackwell, 2003), 63–89.

Zhu, Ning, ‘‘The Local Bias of Individual Investors,’’ Mimeo, Yale, 2002.


Dow




ber 2019

Documents

I. Introductionweb.stanford.edu/~saumitra/papers/JhaQJE2015.pdfSaumitra Jha The English Parliament’s struggle for supremacy against monarchical dic-tatorship during the Civil War