Money Before Banks: Reconstructing London’s Credit Architecture, c. 1340–1560
A new research programme combining monetary history, legal history, institutional economics, archival evidence, and locally trained manuscript-recognition models
Keywords: medieval London; money; credit; bullion famine; guilds; livery companies; debt; trust; monetary institutions; legal history; economic history; handwritten text recognition; local language models; Bank of England
Thesis statement: London did not wait for incorporated banks to become a sophisticated financial centre. Between the mid-fourteenth and mid-sixteenth centuries, merchants, guilds, city courts, royal institutions, and customary legal practices created a dense architecture of credit that was repeatedly tested by monetary scarcity, debasement, war, plague, and trade disruption. My research asks whether those shocks merely changed the quantity of credit or, more importantly, changed its institutional form: who extended it, how obligations were secured, which forums enforced them, and how trust was maintained when coined money was scarce or unstable. The project does not assume a straight line from medieval London to the Bank of England. It asks instead whether two centuries of recurrent monetary stress generated a repertoire of legal, commercial, and associational solutions that later generations could reuse, modify, or reject.
I am beginning a new research programme at Birkbeck, University of London, on the history of English money, credit, and financial institutions. The formal period is approximately 1340 to 1560. The geographical centre is London. The broader problem is simple to state but difficult to answer: how did London function as an international financial centre long before the creation of incorporated public banking institutions?
The question matters because the familiar history of English finance is often told from the late seventeenth century backwards. The Bank of England appears in 1694. The funded national debt, transferable public securities, the growth of organised financial markets, and the institutional changes associated with the so-called Financial Revolution then dominate the story. That literature is indispensable. Dickson (1967), Carruthers (1996), Murphy (2009), and Wennerlind (2011), among others, explain how political institutions, public credit, markets, and ideas about money were transformed in the seventeenth and early eighteenth centuries.
But London had already been conducting large-scale domestic and international commerce for centuries. Merchants bought and sold on credit. Debts were recorded, transferred, guaranteed, disputed, settled, arbitrated, and enforced. Bills of exchange moved value across borders. Recognisances and bonds formalised obligations. Guilds and livery companies created dense networks of reputation and discipline. City courts operated alongside common-law courts and statutory mechanisms. Royal monetary policy altered the quality and availability of coin. Foreign merchants brought their own practices and institutional relationships into the city. None of this requires us to pretend that medieval London had a modern banking system. It requires us to take seriously the sophistication of financial practice before the later institutions that now dominate the historical vocabulary.
That is the starting point of the project.
The historical problem: money can become scarce without exchange disappearing
The period contains two major episodes conventionally described as bullion famines: approximately 1370–1420 and 1440–1480. The precise chronology, severity, causes, and even interpretation of these contractions have been debated extensively. Day’s (1978) classic account made the fifteenth-century bullion famine a major explanatory problem. Munro’s work connected bullion flows, trade, monetary policy, and the Anglo-Burgundian commercial relationship. Spufford (1988) placed late-medieval monetary scarcity within the wider European monetary system. Sussman (1998) challenged important parts of the conventional bullion-famine interpretation. Bolton (2012) and Allen (2012) provided modern syntheses of English monetary conditions, minting, circulation, and institutional structure.
What interests me most is what happens one level below those aggregate monetary measures.
Suppose mint output falls. Suppose silver becomes scarce. Suppose monetary contraction contributes to falling prices, tighter settlement conditions, or changes in trade. What happens to the actual mechanisms by which merchants continue doing business?
One possibility is simple substitution: less coin produces more credit. But that may be too crude. Credit itself has structure. A merchant may become more willing to transact with someone inside a known guild network and less willing to extend unsecured credit to an outsider. A lender may demand a recognisance where an informal promise previously sufficed. Parties may shift from one court to another. They may use sureties more frequently. They may shorten maturity. They may rely on arbitration rather than litigation. A guild may become more important as an information and enforcement institution. A statutory debt-registration system may become more attractive because it changes the expected cost of default.
The project therefore distinguishes monetary quantity from institutional response.
A useful warning can be expressed very simply:
Observed litigationₜ ≠ total credit activityₜ
A rise in debt litigation might reflect more credit, but it might instead reflect more default, weaker informal enforcement, a change in procedural incentives, or movement into a particular jurisdiction. A decline might indicate less credit, but it could equally reflect more successful arbitration, more internal guild enforcement, different documentary practices, or a shift to another court.
For that reason I am treating the credit system as several observable layers rather than one variable:
Credit systemₜ = creationₜ + operationₜ + enforcementₜ
The archival problem is then to find sources that reveal each layer independently enough to prevent one from being mistaken for the whole.
London as an institutional system
The city itself has to be understood before the transactions can be counted. This is why Caroline Barron’s London in the Later Middle Ages is important to the project. Barron (2004) reconstructs the relationship between city government, the Crown, economic infrastructure, trade, courts, elected office, and urban self-government. Sylvia Thrupp’s older The Merchant Class of Medieval London remains valuable because it examines the composition, social structure, political position, and institutional life of the mercantile community itself (Thrupp, 1969).
These works matter because “the market” cannot be treated as an institutionless space in which anonymous buyers and sellers simply meet. Medieval and Tudor commerce was embedded in legal status, citizenship, guild membership, reputation, kinship, partnership, office holding, geography, and repeated dealing. Greif (2006) provides one important institutional framework for thinking about commitment and enforcement in medieval trade. Ogilvie (2011) provides a necessary counterweight by asking when guild institutions facilitated commerce and when privilege, exclusion, and rent extraction mattered more. The project therefore does not begin with the assumption that guilds were either efficient substitutes for markets or obstacles to them. That has to be demonstrated from the evidence.
The same caution applies to “trust”. I am not using trust as a decorative historical word. I want to make it observable where possible. A repeated surety relationship is evidence. A merchant who appears repeatedly with the same counterparties is evidence. Guild membership is evidence of one institutional relationship. Arbitration by wardens is evidence of another. A recognisance enrolled before an authority is a different mechanism again. Trust can therefore be decomposed into practices that reduce information, commitment, and enforcement problems.
At a conceptual level, the question becomes:
Δ institutional form = f(monetary stress, legal options, network structure, information, enforcement costs)
This is not an econometric claim imposed on the sources. It is a way of asking disciplined historical questions. If the evidence contradicts the framework, the evidence wins.
The legal architecture is part of the economy
One of the most important developments in the project has been recognising that legal history cannot be treated as a supporting appendix to monetary history. The legal architecture is part of the economic mechanism itself.
Penny Tucker’s Law Courts and Lawyers in the City of London, 1300–1550 is therefore becoming foundational reading (Tucker, 2007). London’s courts did not merely process disputes after economic relationships failed. Their procedures affected the expected enforceability of obligations before contracts were made. The availability of particular remedies, differences between city custom and common law, the speed and cost of proceedings, opportunities for representation, evidentiary practices, and the possibility of arbitration all influenced the structure of commercial relationships.
The Court of Common Pleas is especially important. If London merchants could choose between city institutions and a central common-law forum, changes in where cases appear may reveal strategic forum choice rather than changes in underlying credit volume. This makes the institutional history of the courts indispensable before quantitative claims are made from litigation data.
The same logic applies to Statute Merchant and Statute Staple procedures. Goddard (2016) shows the importance of the Statute Staple as a debt-registration and enforcement mechanism for later-medieval trade finance. Such records potentially allow the project to observe formal credit creation rather than merely the disputes that eventually reached litigation.
The empirical design therefore has to distinguish at least three things:-
Credit creation: recognisances, registered obligations, bonds, bills, and other formal instruments.
-
Credit operation: merchant accounts, guild accounts, bills of exchange, ongoing balances, repeated transactions, and successful settlement.
-
Credit enforcement: litigation, default proceedings, arbitration, guild discipline, attachment, judgment, settlement, and withdrawal.
This is one of the central methodological commitments of the research.
The primary sources
The project begins with the London Plea and Memoranda Rolls, preserved in calendar form for 1323–1482 and supplemented by unedited archival material. These records document debt litigation, recognisance enrolments, commercial obligations, settlement, and the procedural environment of London’s courts. Rather than reading them only for notable anecdotes, I intend to construct sampled transaction-level datasets across defined periods.
For each transaction, the goal is to record as much as the evidence permits: date, parties, occupation, origin, guild affiliation, obligation type, value, maturity, security, surety, documentary form, forum, procedural action, outcome, arbitration, settlement, repeated dealing, and links to other people or institutions.
The first article will probably begin with a deliberately limited set of years. A baseline before severe monetary stress will be compared with years during the first bullion famine. The purpose is not immediately to maximise sample size. It is to establish whether the variables survive consistently enough to support a much larger study.
The second major source family is the records of London’s guilds and livery companies. Merchant Taylors’, Goldsmiths’, Mercers’, and other company records preserve forms of evidence that courts cannot. Court records disproportionately observe disputes and enforcement. Company accounts and minutes can reveal ongoing governance, internal discipline, surety, debt, arrears, payments, arbitration, corporate intervention, and repeated relationships.
Davies and Saunders (2004) provide an institutional history of the Merchant Taylors’ Company that will help situate those records. Nightingale’s work on the Grocers’ Company and mercantile credit demonstrates how company archives can illuminate monetary and commercial behaviour that is invisible in aggregate mint data (Nightingale, 1990, 1995).
The Goldsmiths’ material is particularly attractive because the surviving digitised manuscript sequence reaches back into the 1330s. That creates the possibility of observing one institutional tradition before, during, and after the first major period of bullion scarcity.
Figure 1. Example manuscript page from the Goldsmiths’ Company wardens’ accounts/court-minute material currently being examined. The project preserves the manuscript image as the controlling source and separates machine transcription, human correction, normalisation, translation, and structured extraction into distinct layers.
The Court of Common Pleas will provide an external legal comparison. Statute Staple records will provide another view of formal credit and default. City journals, Letter Books, Close Rolls, Patent Rolls, and royal monetary legislation will help reconstruct the policy environment. Exchequer and mint records provide the macroeconomic monetary context. Coin-hoard and numismatic evidence matter because mint output is a flow, not a complete measure of the stock or composition of money in circulation.
Merchant evidence offers yet another observational layer. The surviving Borromei Bank London ledger from the 1430s, for example, is valuable precisely because merchant books capture functioning commercial relationships that may never appear in litigation. De Roover’s (1948) study of Bruges remains useful comparatively because it demonstrates the range of banking, exchange, deposit, credit, and bullion practices operating in another major late-medieval commercial centre.
The project is therefore intentionally plural in its evidence. No single archive will be treated as “the credit system”.
What I have started doing
The first stage is already under way. I have begun working through digitised livery-company material available through the University of London and Senate House environment, including Goldsmiths’ Company wardens’ accounts and court-minute material. The pages are immediately revealing for two reasons.
First, they contain exactly the institutional evidence the project needs: obligations, wardens, disputes, collective decisions, payments, office holding, and the governance of the craft. Second, they demonstrate why historical language and handwriting are themselves part of the research problem.
The manuscripts do not arrive as modern prose waiting to be searched. They contain heavily abbreviated documentary hands, Anglo-Norman/French, Latin, changing orthography, specialist institutional vocabulary, proper names, numerical notation, marginal entries, and scribal variation. A historian can read such material manually, but at scale the bottleneck becomes severe.
My previous medieval-history training included working with Latin documentary sources. I am now combining that palaeographic and linguistic work with my computing background to build a local document-analysis environment specifically for this corpus.
The important word is local.
I do not want a general-purpose model to act as an oracle. I want a controlled system that can be trained on the exact manuscript traditions being studied, corrected repeatedly, versioned, tested against held-out material, and audited back to the page image.
The local LLM and handwriting-recognition programme
Handwritten Text Recognition is already a serious field. Platforms such as Transkribus demonstrated that neural recognition systems could make large historical manuscript collections searchable and transcribable when suitable ground truth is available (Kahle et al., 2017; Mühlberger et al., 2019). My interest is not in reproducing that platform. It is in building a research-specific pipeline around local models, expert correction, historical-language interpretation, and structured economic extraction.
The basic pipeline is:
manuscript image → layout/line detection → initial HTR → local language-model correction → human palaeographic review → diplomatic transcription → abbreviation expansion → normalised text → translation → named entities → transaction extraction → historical database
Each arrow is a separate stage. Nothing is allowed to overwrite what came before it.
That matters because the final database will contain claims about people, obligations, amounts, dates, legal forms, and institutional relationships. If a later interpretation proves wrong, I need to be able to move backwards through the chain and determine whether the error arose in image recognition, abbreviation expansion, translation, entity resolution, or historical classification.
The original page remains the authority.
Figure 2. A second example from the manuscript corpus. Entries combine narrative institutional material with names, accounts, sums, and abbreviated documentary language. These are precisely the conditions under which expert-corrected, collection-specific training data become more valuable than generic OCR.
The first training corpus will therefore be small by machine-learning standards and high quality by historical standards. I would rather have hundreds of pages with carefully verified transcriptions than tens of thousands of pages containing silent errors. The initial model should learn the hands, abbreviations, formulae, vocabulary, and layout conventions of a constrained source family. Once the system performs reliably on held-out pages from that family, the corpus can expand.
For every page, I intend to retain at least the following layers:-
original image;
-
archive and document reference;
-
date or date range;
-
raw HTR output;
-
local-model revision;
-
corrected diplomatic transcription;
-
expanded abbreviations where appropriate;
-
normalised reading;
-
translation;
-
uncertain readings and confidence notes;
-
named entities;
-
structured transaction data;
-
model and prompt/version metadata;
-
a cryptographic hash or equivalent integrity marker for the source and derived files.
The human correction is not an inconvenience to be minimised. It is the most valuable training data in the system. Each corrected line improves the model’s understanding of the collection and creates a documented relationship between machine output and expert judgement.
Why this matters for historical method
There is an obvious temptation with artificial intelligence in archival research: process more material, faster, and then celebrate scale. I think that is the wrong objective.
Scale is useful only if evidential discipline scales with it.
A million machine-generated transcriptions are not automatically better than ten thousand carefully controlled ones. A model can produce plausible expansions of abbreviations that are wrong. It can silently modernise spelling. It can turn an unfamiliar name into a familiar one. It can infer a legal relationship that the source does not state. It can make an uncertain line sound certain. These are not merely technical errors. They can alter the historical argument.
The system therefore needs explicit uncertainty. If a word has two plausible readings, both should survive until resolved. If a name cannot be identified confidently, it should remain unresolved rather than being forced into an entity table. If an amount is ambiguous, the ambiguity must propagate into any quantitative analysis.
This is also why I am separating transcription from translation and translation from interpretation. A diplomatic transcription answers one question: what marks on the page are being read as what text? Expansion answers another: what do the abbreviations represent? Translation is another act again. Historical classification — deciding, for example, that a passage records a surety relationship or an arbitrated debt — is a further interpretive layer.
Collapsing those stages would make the process faster and the evidence weaker.
The first substantive paper: bullion famine and institutional adaptation
The first publication I plan to develop from the project will test whether the composition and enforcement of credit changed when metallic money became scarce.
The obvious hypothesis is that credit substitutes for money. But the more interesting hypotheses concern the structure of that substitution.
During monetary stress, do merchants rely more heavily on known counterparties? Does guild affiliation become more predictive of repeated transactions or surety? Do formal recognisances become more common relative to less formal obligations? Are maturities shortened? Are different courts used? Do wardens or corporate arbitration appear more frequently? Does the value distribution of recorded debts change? Do creditors become more selective? Does enforcement migrate between city courts, Common Pleas, guild mechanisms, and statutory debt procedures?
Nightingale’s (1990) work on monetary contraction and mercantile credit is central to this question. Goddard (2016) provides a broader analysis of credit and trade across later-medieval England, including Statute Staple evidence and London’s role as a commercial centre. Day (1978), Munro (1983), Sussman (1998), Bolton (2012), Allen (2012), and Spufford (1988) provide competing ways of defining and interpreting the monetary shock itself.
The article will therefore have to do something that is methodologically simple but often neglected: keep the independent and dependent evidence separate. Monetary conditions should be constructed from monetary and numismatic evidence. Credit behaviour should be reconstructed from transaction and institutional records. The two should then be compared. The existence of a “bullion famine” should not be inferred from the same credit evidence later used to show its effects.
The strongest version of the paper would not claim merely that credit increased when coin became scarce. It would show that monetary stress altered the institutional architecture of exchange.
That is a much more interesting historical claim.
A wider thesis about institutional repertoires
The full PhD extends beyond the first bullion famine. The second major period of scarcity, approximately 1440–1480, may prove even more important because it allows comparison between repeated shocks. The Tudor transition then provides a period in which practices developed under stress can be examined under different monetary conditions. Finally, the Great Debasement of the 1540s creates a different kind of monetary shock.
That difference matters. A shortage of coin is not the same problem as degradation in the quality of coin. The first constrains liquidity. The second can alter confidence in the medium itself, relative prices between denominations, hoarding incentives, payment preferences, and the terms on which nominal obligations are accepted.
If London’s institutions respond differently to scarcity and debasement, that distinction can tell us something important about what “monetary trust” actually meant in practice.
The broader thesis is not that these medieval and Tudor institutions inevitably produced the Bank of England. That would be teleological and, in my view, historically weak. The claim I want to test is more restrained and more defensible: repeated crises may have produced a repertoire of enforceable practices — recognisances, surety structures, corporate arbitration, bills of exchange, documentary conventions, procedural remedies, and institutional relationships — that later economic actors inherited as part of the available institutional environment.
Institutions rarely emerge from nothing. But neither are they predetermined by what came before.
That distinction between inheritance and inevitability will run through the project.
Reading the historiography forwards and backwards
The reading programme is deliberately chronological as well as thematic. Older work matters because later scholars often inherit the categories of earlier debates even when they reject the conclusions.
Postan’s early work on medieval trade and credit helped establish credit as fundamental to medieval commerce. The later “bullion famine” and contraction debates then connected monetary supply to wider economic performance. Day (1978) gave the fifteenth-century famine one of its classic formulations. Munro placed bullion, monetary policy, and international trade together. Nightingale brought mercantile credit directly into the monetary-contraction problem. Sussman reconsidered the macroeconomic logic. Bolton and Allen subsequently strengthened the monetary and numismatic foundations.
Alongside that literature, Barron, Thrupp, Tucker, Davies, Goddard, Greif, and Ogilvie force the analysis back into institutions: city government, merchants, courts, companies, enforcement, information, privilege, and reputation.
The point is not to choose one literature and append the other. The project exists because the macro-monetary and micro-institutional literatures do not fully answer the same question.
Mint output can tell us that fewer coins were being produced. It cannot tell us whether a Goldsmiths’ Company warden intervened in an obligation between two members. A court roll can tell us that a debt was litigated. It cannot, by itself, tell us the size of the circulating money stock. A guild minute can reveal arbitration. It cannot establish the national bullion balance. A merchant ledger can reveal functioning credit but may tell us little about disputes settled elsewhere.
The research design therefore treats disagreement between source types as evidence to be explained, not noise to be averaged away.
What success would look like
The project will succeed if it produces three things.
First, it should make a substantive contribution to the history of money and credit by reconstructing how London’s credit institutions actually operated across repeated monetary shocks.
Second, it should produce a source-linked transaction dataset in which every observation can be traced back to an archival record rather than existing as an opaque statistical row.
Third, the manuscript work should produce a reproducible method for using locally trained HTR and language models in historical research without allowing machine output to become detached from palaeographic judgement.
The computational method is not the thesis. The history is the thesis. But better tools can change what historical questions are empirically possible.
A historian who can reliably process a few hundred pages can ask one class of question. A historian who can process tens of thousands of pages while retaining line-level provenance can ask another. Network structures that were previously anecdotal can become measurable. Repeated counterparties can be identified across decades. Changes in legal instruments can be quantified. Institutional migration between forums can be observed. Rare events can be found without relying on indexes compiled for different purposes.
The danger is obvious: a machine can make the archive look cleaner than it is. My aim is the opposite. The system should preserve the archive’s uncertainty while making more of it analysable.
Where the project stands now
At this stage I am doing three things in parallel.
The first is historical: working through the core monetary, urban, legal, and credit historiography, including Barron, Thrupp, Tucker, Nightingale, Day, Munro, Spufford, Sussman, Bolton, Allen, Goddard, Greif, and Ogilvie.
The second is archival: identifying the first controlled sample from the Plea and Memoranda Rolls and overlapping livery-company material, particularly the Goldsmiths’ records, before scaling into Common Pleas, Statute Staple, merchant, City, royal, and numismatic sources.
The third is computational: constructing the local HTR/LLM environment, defining the data model, preserving provenance, and beginning the expert-corrected training corpus from manuscript pages whose institutional context is known.
The next milestone is not “big data”. It is a small, complete evidential chain: one historical episode seen simultaneously through monetary evidence, court transactions, institutional records, named people and networks, and the original manuscript page.
If that works, scale becomes an engineering problem rather than a methodological gamble.
And that is where the project begins: not with the Bank, but with the obligations, accounts, courts, companies, coins, and handwritten pages that made financial life possible before it.
References
Allen, M. (2012). Mints and money in medieval England. Cambridge University Press.
Barron, C. M. (2004). London in the later Middle Ages: Government and people 1200–1500. Oxford University Press. https://doi.org/10.1093/acprof:oso/9780199257775.001.0001
Bolton, J. L. (2012). Money in the medieval English economy, 973–1489. Manchester University Press.
Carruthers, B. G. (1996). City of capital: Politics and markets in the English financial revolution. Princeton University Press.
Davies, M. (Ed.). (2000). The Merchant Taylors’ Company of London: Court minutes, 1486–1493. Paul Watkins.
Davies, M., & Saunders, A. (2004). The history of the Merchant Taylors’ Company. Maney.
Day, J. (1978). The great bullion famine of the fifteenth century. Past & Present, 79(1), 3–54. https://doi.org/10.1093/past/79.1.3
De Roover, R. (1948). Money, banking and credit in mediaeval Bruges: Italian merchant-bankers, Lombards and money-changers. Mediaeval Academy of America.
Dickson, P. G. M. (1967). The financial revolution in England: A study in the development of public credit, 1688–1756. Macmillan.
Goddard, R. (2016). Credit and trade in later medieval England, 1353–1532. Palgrave Macmillan. https://doi.org/10.1057/978-1-137-48987-6
Greif, A. (2006). Institutions and the path to the modern economy: Lessons from medieval trade. Cambridge University Press.
Kahle, P., Colutto, S., Hackl, G., & Mühlberger, G. (2017). Transkribus—A service platform for transcription, recognition and retrieval of historical documents. In 2017 14th IAPR International Conference on Document Analysis and Recognition (ICDAR) (pp. 19–24). IEEE. https://doi.org/10.1109/ICDAR.2017.307
Mühlberger, G., Seaward, L., Terras, M., Ares Oliveira, S., Bosch, V., Bryan, M., Colutto, S., Déjean, H., Diem, M., Fiel, S., Gatos, B., Greinoecker, A., Grüning, T., Hackl, G., Haukkovaara, V., Heyer, G., Hirvonen, L., Hodel, T., Jokinen, M., … Zagoris, K. (2019). Transforming scholarship in the archives through handwritten text recognition: Transkribus as a case study. Journal of Documentation, 75(5), 954–976. https://doi.org/10.1108/JD-07-2018-0114
Muldrew, C. (1998). The economy of obligation: The culture of credit and social relations in early modern England. Macmillan.
Munro, J. H. (1983). Bullion flows and monetary contraction in late-medieval England and the Low Countries. In J. F. Richards (Ed.), Precious metals in the later medieval and early modern worlds (pp. 97–158). Carolina Academic Press.
Murphy, A. L. (2009). The origins of English financial markets: Investment and speculation before the South Sea Bubble. Cambridge University Press.
Nightingale, P. (1990). Monetary contraction and mercantile credit in later medieval England. The Economic History Review, 43(4), 560–575. https://doi.org/10.1111/j.1468-0289.1990.tb00545.x
Nightingale, P. (1995). A medieval mercantile community: The Grocers’ Company and the politics and trade of London, 1000–1485. Yale University Press.
Ogilvie, S. (2011). Institutions and European trade: Merchant guilds, 1000–1800. Cambridge University Press. https://doi.org/10.1017/CBO9780511974410
Spufford, P. (1988). Money and its use in medieval Europe. Cambridge University Press. https://doi.org/10.1017/CBO9780511583544
Sussman, N. (1998). The late medieval bullion famine reconsidered. The Journal of Economic History, 58(1), 126–154. https://doi.org/10.1017/S0022050700019914
Thomas, A. H., & Jones, P. E. (Eds.). (1926–1961). Calendar of Plea and Memoranda Rolls preserved among the archives of the Corporation of the City of London at the Guildhall (6 vols.). Cambridge University Press.
Thrupp, S. L. (1969). The merchant class of medieval London, 1300–1500. University of Michigan Press. (Original work published 1948)
Tucker, P. (2007). Law courts and lawyers in the City of London, 1300–1550. Cambridge University Press. https://doi.org/10.1017/CBO9780511585845
Wennerlind, C. (2011). Casualties of credit: The English financial revolution, 1620–1720. Harvard University Press. https://doi.org/10.4159/harvard.9780674062665