Data Sources

Global Level

Indicators of Globalization, 1980-2004 By Mauro Guillen.

To track changes in the degree of globalization.

Variables: A. Economic:
Exports + imports of goods & services, % world GDP
Developed countries, % GDP
Developing countries, % GDP
Inward foreign direct investment stock, % world GDP
Developed countries, % GDP
Developing countries, % GDP
Exports of foreign affiliates, % total world exports

B. Financial:
Daily currency exchange turnover, % world GDP
Cross-border bank loan stock, % world GDP
Cross-border banking claims stock, % world GDP

C. Internet:
Users, % world population:
Developed countries, % population
Developing countries, % population

D. Demographic:
Stock of international migrants, % world population
Refugee population, % world population

E. Political:
Nation-states with membership in the United Nations
International organizations, number:
Intergovernmental
Non-governmental

F. Social & Cultural

International tourist arrivals, % world population
International calls, minutes per capita
Books in Print on the topic of globalization

New Maddison Historical Dataset of Country GDPs, Year 1 until 2010. By Agnus Maddison, Jutta Bolt and Jan Luiten van Zanden.

The Maddison Project, initiated in March 2010 by a group of close colleagues of Angus Maddison, aims to develop an effective way of cooperation between scholars to continue Maddison’s work on measuring economic performance in the world economy. This paper is a first product of the project. Its goal is to inventory recent research on historical national accounts, to briefly discuss some of the problems related to these historical statistics and to extend and where necessary revise the estimates published by Maddison in his recent overviews (2001; 2003; 2007)

Variables: Country. Year. GDP, per capita.

Country Level

Index of Central Bank Independence, 1989-2000 By Simone Polillo and Mauro F. Guillén.

To track the changes in the degree of central bank independence

Variables: Country
Year
Central Bank Indepenence

Historial Statistics of the United States – Online By Susan Carter, Scott Gartner, Michael Haines, Alan Olmstead, Richard Sutch, and Gavin Wright (Editors).

To provide a definitive source for American statistical information, containing over 37,000 times series and 70 original essays on
the history and sources of data.

Variables: Any conceivable variable.

Patent Citations and the Geography of Knowledge Spillovers: A Reasssessment By Peter Thompson & Melanie Fox Kean.

Variables: The data used in the paper is in a zipped excel file. Variable definitions are included in the first sheet of the file. Geographic correlation files used to translate inventor locations to metropolitan statistical areas have been superseded by correlation files available here.

Patent Citations and the Geography of Knowledge Spillovers By Peter Thompson.

Variables: This tab-delimited file contains the data used in the paper. It consists of all patents with an insutitutional assignee granted during the first week of January, 2003, and all patents they cite granted after January 1, 1976. The citing patents appear mutliple times, once for each citation they make.
The columnns of the files are:

Citing number- Number of the citing patent.
Cited number- Number of the cited patent.
Added by examiner Equals 1 if the examiner added the citation, 0 otherwise
Cited noninstitutional Equals 1 if the assignee of the cited patent is a company, university, or government entity, 0 if the assignee is a private individual or no assignee other than the inventor is given.
Self-citation Equals 1 if the citation is to a patent with the same assignee.
*Citing file date Filing date of citing patent.
*Citing issue date- Issue date of citing patent.
*Citing assignee name- In the case of multiple assignees, the first-named assignee was used. Where possible, the file name was converted to the preferred format given in the NBER patent files. This variable needs further manual cleaning before it can be used reliably.
*Citing assignee city- The city of the assignee as given in the patent. I did not use this variable in the paper, so it has not been cleaned up. Some place names are missing, and foreign place names (especially Taiwanese) have all sorts of variants of spelling. This variable needs further manual cleaning before it can be used reliably.
*Citing assignee state- The state of the assignee as given in the patent. I did not use this variable in the paper, so it has not been cleaned up. In particular, the state code often confounds countries and US states. For example, DE is Delaware and Germany, CA is California and Canada. It is tedious to clean these up by hand, but it can be done. While the html version of the patent on the USPTO web site does not clarify, the image file does. Until 2001, for example, the patent image shows “Calif.” and “Canada” while the html file uses “CA” for both. After January 2001, the image gives “CA (US)” and “CA” for the state and county resp[ectively. This variable needs further manual cleaning before it can be used reliably.
*Citing first inventor city The city of residence of the first-named inventor. Please note that this variable has not been cleaned up for foreign locations. Foreign place names (especially Taiwanese) have all sorts of variants of spelling, and often other parts of a mailing address are still included.
*Citing first inventor state- The state of residence of the first-named inventor. Noted as “Foreign” if non-US.
*Citing first inventor county- The county of residence of the first-named inventor. Noted as “Foreign” if non-US. See notes to correlation file above.
*Citing first inventor msa- The MSA of residence of the first-named inventor. Noted as “Foreign” if non-US. See notes to correlation file above.
*Citing first inventor cmsa- The CMSA of residence of the first-named inventor. Noted as “Foreign” if non-US. See notes to correlation file above.
*Citing first inventor country- The country of residence of the first-named inventor.
*Citing examiner- The name of the primary examiner.
*Citing US class- The primary US classification code.
*Citing Int class- The International classification code. Where more than one code is given in the patent, the first-listed is given here.

Multiple Firms & Industry Level

U.S. Tire Firms, 1920-1981 By Steven Klepper and Kenneth L. Simons.

To study the evolution of the industry of the industrial structure of the U.S. tire Industry.

Variables: firmname – Name (subsequent names after semicolons) to 80 characters
backdate – First year listed in Hendricks if earlier than in Thomas’, else 0
entryyr – First year listed in Thomas’
exityr – First year not listed in Thomas’ (1981 signifies still a producer at end of sample)
acquired – 1 if acquired rather than exited, 0 otherwise
trsize20 – its size code in the 1920 Thomas’ volume (1000, 750, 400, 200, 75, 37.5, 17.5, and .01 which correspond to the letter codes AAAA, AAA, AA, A, B, C, D, and X, with the five largest 1920 firms, Goodyear, Goodrich, Firestone, US Rubber, and Fisk coded as 2000)
akron – 1 if located within 50 miles of Akron, 0 otherwise
cord1917 – 1 if produced a cord tire in the 1917 Tire Rate Book, 0 otherwise
cord1920 – 1 if produced a cord tire in the Oct. 1920 Tire Rate Book, 0 otherwise
ball1923 – 1 if produced a balloon tire in the Dec. 1923 Tire Rate Book, 0 otherwise
distms – 1 if had dealers and/or distributors in multiple states, 0 if not, 9 if unknown

Working paper / publications: Steven Klepper and Kenneth L. Simons entitled, “The Making of an Oligopoly: Firm Survival and Technological Change in the Evolution of the U.S. Tire Industry,” Journal of Political Economy, 2000, vol. 108, no. 4, pp. 728-760.

Steve here you can put some notes.

U.S. Television Receiver Producers, 1947-1990 By Steven Klepper and Ken L. Simons.

Study the evolution of market structure in the U.S Television Receiver Industry with particular focus on the impact that prior experience in radio manufacturing had on the life chances of TV receiver producers.

Variables: firmname – Name (subsequent names after semicolons) to 80 characters
entryyr – Year of entry into tv manufacture according to Television Factbook
exityr – First year no longer producing tvs according to Television Factbook, for entrants by 1951 (otherwise blank)
exited – Whether exited versus still producing at end of sample or exited by acquisition by another television firm, 1 if true 0 otherwise
radio – Whether previously produced radios according to Thomas’, 1 if true 0 otherwise
homerad – Whether previously produced home-type radios according to Thomas’, 1 if true 0 otherwise
unkntype – Whether previously produced unknown-type radios according to Thomas’, 1 if true 0 otherwise
radtvsz – Numeric code corresponding to the sizefl variable in the radios.txt dataset, and ranging from 0 (unknown size) to 12 (largest), blank if not relevant
yrsexp – Years of radio experience
wilrandd – Willard’s R&D rating, blank if not available
wildists – Willard’s distribution rating, blank if not available
lms52 – Log of market share in 1951-53, blank if not available
lms60 – Log of market share in 1959-60, blank if not available
lms70 – Log of market share in 1970, blank if not available
firmpdin – Product innovations per year as used in table 8, blank if firm was not producing during years from which data were collected
firmpcin – Process innovations per year as used in table 8, blank as above
firmin1 – Total innovations per year over years preceding 1951, blank as above
firmin2 – Total innovations per year over years preceding 1959, blank as above
firmin3 – Total innovations per year over years preceding 1970, blank as above

Working paper / publications: Steven Klepper and Kenneth L. Simons, “Dominance by Birthright: Entry of Prior Radio Producers and Competitive Ramifications in the U.S. Television Receiver Industry,” Strategic Management Journal, vol. 21, pp. 997-1016

The two data sets contain observations on radio producers at the start of the television industry and television producers that began producing televisions by 1951, where the set of radio producers and the set of television producers and the variables in the two data files are explained in detail in Steven Klepper and Kenneth L. Simons, “Dominance by Birthright: Entry of Prior Radio Producers and Competitive Ramifications in the U.S. Television Receiver Industry,” Strategic Management Journal, vol. 21, pp. 997-1016. The data in the dataset is organized in a tab-delimited ASCII file.

U.S. Radio Receiver Producers, 1951 By Steven Klepper and Ken L. Simons.

Study what kind of radio producers entered TV receiver industry.

Variables: firmname – Name (subsequent names after semicolons) to 80 characters
sizefl – Size code in Thomas’ Register when first listed
homerad – Whether indicated as a home radio type producer, 1 if true else 0
unkntype – Whether indicated as an unknown-type radio type producer, 1 if true else 0
yrsexp – Years of radio experience entertv – Whether entered TV production, 1 if true 0 if not, deliberately missing (blank) for Raytheon see footnote 7 in the paper.

The two data sets contain observations on radio producers at the start of the television industry and television producers that began producing televisions by 1951, where the set of radio producers and the set of television producers and the variables in the two data files are explained in detail in Steven Klepper and Kenneth L. Simons, “Dominance by Birthright: Entry of Prior Radio Producers and Competitive Ramifications in the U.S. Television Receiver Industry,” Strategic Management Journal, vol. 21, pp. 997-1016. The data in each of the two data sets are organized in a tab-delimited ASCII file.

IEHA data hub “Firms and capital markets” By Joerg Baten and colleagues, plus links to data by W. Goetzmann, K. Geert Rouwenhorst , Christos Cabolis, William Goetzmann, Peter Radchenko, Roger G. Ibbotson, K. Geert Rouwenhorst, Wenzhong Fan, Fred Bateman and Liang Peng, Jeremy Atack, Fred Bateman and Thomas Weiss.

Data on firms and capital markets are one of the most important sources of information in and economic-historical research, especially for the pre-statistical period. Scholars working in this field are invited to register their work, and to make their data bases available through the internet – either on their own webpages (which may be made accessible via this list) or on our webpages. Scholars like Christos Cabolis, William Goetzmann, Peter Radchenko, Roger G. Ibbotson, K. Geert Rouwenhorst, Wenzhong Fan, Jeremy Atack, Thomas Weiss, Fred Bateman and Liang Peng make their data bases available through the internet, especially on our homepage. If you are interested to make your data set available on our homepage, it is also possible to send the data set with the obligation that we make it available only in two or three years after the first major publication. The long term goal is to enhance the exchange and (re-)use of these data in order to write truly international-comparative histories of the development of firms and capital markets and their consequences. All data bases will not only include the data themselves, but also descriptions about the data.

Data base login USA
(Explanation and hints)

You have to be a registered user to download the files of Christos Cabolis, William Goetzmann, Peter Radchenko, Roger G. Ibbotson, K. Geert Rouwenhorst, Wenzhong Fan and Liang Peng . If you have not registered before, click here to register now. If you are a registered user, login to download the files. They will email you the password.Passwords are not encrypted, please do not register your regular password for this system. The information gathered from the form will be held in strictest confidence.

Index to Datafiles
GERMANY

Profits and capital stock informationfrom tax-lists . Germany (1878-1913)
– Author: Jörg Baten
– About the pilot project
– The datafile : Data sets (zipped)

RUSSIA

St. Petersburg Stock Exchange
– Author: Christos Cabolis, William Goetzmann, and Peter Radchenko
– About the St. Petersburg Stock Exchange1865-1917
– The datafile : Data sets (zipped)

USA
New York Stock Exchange
– Author: William N. Goetzmann, Roger G. Ibbotson and Liang Peng
– About the New York Stock Exchange 1815-1925
– The datafile : Data sets (zipped) (scanned data, has to be transferred to Excel)

Documentation on coding in the Bateman-Weiss and Atack-Bateman sampels from the manuscript censuses of manufactures
– Author: Jeremy Atack, Fred Bateman and Thomas Weiss
– About the data set 1850-1880
– The datafile: Data sets

ENGLAND

London Stock Exchange
– Author: William N. Goetzmann, K. Geert Rouwenhorst
– About the London Stock Exchange 1869-1929
– The datafile : Data sets (zipped)

CHINA

Shanghai Stock Exchange
– Author: Christos Cabolis, Wenzhong Fan and William Goetzmann
– About the Shanghai Stock Exchange 1870-1940
– The datafile: Data sets (zipped)

Variables: profit, capital stock

An exploration of learning and spillovers in wartime shipbuilding By Peter Thompson & Rebecca Achee Thornton.

Variables: The data set makes use of a highly detailed set of data collected from various sources. The primary source for this data set was collected from the National Archives. These data were supplemented with information gathered from Ships for Victory by Frederic Lane and A Statistical Summary of Shipbuilding under the U.S. Maritime Commission during World War II by Gerald Fischer.

How much did the Liberty Shipbuilders learn? New evidence for an Old Case Study By Peter Thompson.

Variables: Productivity: Includes USMC hull number; builder’s hull number; vessel’s original name; vessel type; round of way; way number; date of keel laying; launch date; delivery date; production time in days (on ways, in outfitting, and total); labor requirements (direct, indirect and total); unit production costs (yard costs, procurement cost, facilities cost, and administrative cost); fracture class (if any) and date of first major fracture; date of war loss (if applicable). Note that the cost data were produced 1946 to provide benchmarks for post-war auction bids. Total wartime yard expenditure on capital was allocated to each vessel produced according to the time spent in production. The data on facilities costs therefore do not measure contemporaneous flows of capital services used in production.

Employment: Includes monthly data for number of wage earners (total and direct labor); number of salaried employees, total employees; total employees engaged on USMC construction; labor turnover rates (total hires per 100 employees, total separations per 100 employees; separations decomposed into quits, fires, layoffs, military and other reasons); average hours per week worked, wages per hour, absentee rate, distribution of employment by shift (Mon-Fri shifts 1,2,3; Saturday employment and Sunday employment). Researchers can find monthly employment data for the remaining yards in G.J. Fischer, “A Statistical Summary of Shipbuilding Under the US Maritime Commission During World War II,” [Washington DC: Government Printing Office, USMC, 1949], which is easy to obtain on ILL.

Strikes: Each strike listed by start date, end date, duration, number of workers involved, person days lost, BLS code for cause, BLS descriptor for cause
Inventories: Provides monthly estimates of steel plate inventories in tons and ships worth. Even when available for a yard, the data for the period late 1942 to early 1943 are missing. This makes the data all but useless, as it was precisely during this period that steel shortages may have been affecting productivity.

Capital: Provides data on authorizations for capital investment: contract number, date of approval, date of contract, purpose of investment, and amount.

The data are contained in a zipped set of Lotus-123 files (the download link is at the bottom of this page). Each major yard has its own file. Documentation of sources can be found as footnotes to each table.

Technological change and the Age-Earnings Profile: Evidence from the International Merchant Marine By Peter Thompson.

Variables: The data are taken from CDs created by The Maritime History Archive, Memorial University of Newfoundland. The file is a zipped, tab-delimited text file. The first row contains the series names.
WAGE_1900: Monthly wage in constant 1900 G.B. pounds sterling. All mariners in the sample were also given provisions.
AGE: Age in years of the mariner.
BARK: Technology indicator variable equal to 1 if the vessel was a barque or barquentine, and zero otherwise.
STEAM: Technology indicator variable equal to 1 if the vessel was asteamship, and zero otherwise.
Note: If Bark=Steam=0, then the vessel was a square-rigged ship or brig.
CONTRACT_YEAR: Year the mariner embarked on the voyage.
LITERATE: Crude literacy indicator: equal to one if the mariner signed his own name, 0 if a X was used.
COMPLETE: Equal to one if the mariner completed the contract length before leaving the vessel, zero otherwise. Early departure may be the result of desertion or mutual agreement.
CURRENCY: Equal to one if payment was made in US dollars, zero if payment was made in pounds sterling.
GROSS_TONS: Registered gross tonnage of vessel, in thosuands.
YEAR_CONSTRUCTED: Year vessel was launched.
Dummy variables for locations of port of embarkation:
Baseline: Canada
PJ1=1: USA
PJ2=1: Great Britain
PJ3=1: Ireland
PJ4=1: Scandinavia
PJ5=1: Other Europe
PJ6=1: Latin America and Caribbean
PJ7=1: Africa
PJ8=1: Asia/Pacific/Oceania.
Dummy variables for geographic voyage destinations:
Baseline: North Atlantic (specific destination not given — usually multiple European destinations))
V1=1: USA (East Coast, Gulf Coast, and Caribbean)
V2=1: Europe (specific country destination)
V3:=1: South America, USA (west coast)
V4=1: Other (Africa, Asia, Oceania, world voyages)

Selection and Firm Survival. Evidence from the Shipbuilding Industry (1825-1914) By Peter Thompson.

Variables: There are two excel files: firms.xls contains basic information about shipyards; vessels.xls record information about individual vessels. Please remember that this data set covers only vessels with metal hulls with at least 20 tons displacement.
The variables in the files are as follows:

Firms.xls

Key Dates

Founding year
Year company was founded. This can be many years before entry into iron and steel shipbuilding.

Entry year Year first metal vessel was built.
Last year Last year recorded in dataset
Exit year Last year of production in shipbuilding.
Closing year Year company closed. This can be many years after exit from shipbuilding.
Prior Background
A set of dummy variables indicating whether the firm (or its founders) had previously been engaged in any of the following activities:

foundry work, construction, manufacturer of steam engines, locomotive manufacture, railroads, naval engineering, metal shipbuilding (in another firm), wooden shipbuilding, or shipping. These are not mutually exclusive fields.

The category “unknown” arises when I essentially know nothing about the firm.

Takeover of yard
Indicates if firm entered industry by purchasing an existing shipyard engaged in metal shipbuilding.

Firm location
City and state of yard.

Form of Exit
A set of dummy variables indicated why the firm exited:

Bankruptcy, into a trust (such as the American Shipbuilding Corp.), merger, death or retirement of owner, exited industry but continued activities in other areas, firms was sold, a miscellaneous (but known) category, and unknown.

Vessels. xls

The following variables are self-explanatory:
Official number, year built, firm name, vessel type, gross tons, length, breadth, depth, hull material, propulsion system engine, who the vessel was built for, first home port

Year withdrawn and reason withdrawn
This is the year the vessel stops being registered, and the reason for the end of the registration (e.g. vessel was scrapped).

As usual, missing data means I don’t know.

Individual Person Level

Technological change and the Age-Earnings Profile: Evidence from the International Merchant Marine By Peter Thompson.