Article 5 – Inequality of real estate wealth in early modern Stockholm, 1730–1850
4. Sources
Protocols of legal sales
The most important data for this dissertation are the legal sales of buildings in Stockholm from 1730 to 1874. Information about these sales can be drawn from numerous sources, all registered by the administrative body of the city.
The information has been collected and gathered in a database by three colleges and myself.58
Sales of land and buildings have been documented in Sweden for a long time. Both Magnus Eriksson’s city law from 1349 and the old medieval provincial laws (landskapslagarna) establish rules for how these sales are to be processed. By both laws, relatives of the seller had the right to intervene after a transaction was settled and claim their right to buy the property instead.
Not until after the sale was announced (uppbud) three times at the local court (ting), the transaction was sealed.59 The convention to announce real estate sales lived on from medieval times into the 19th century, first to protect family rights and later to tax citizens. The main archival sources are summarized in Table 2 below.
Two archives are of special interest for this thesis and both originates from Stockholm’s city council (Stockholms rådhusrätt): Stockholms magistrat och rådhusrätt 1600-1849 (SMR) and Stockholms rådhusrätt, Avdelning 1 1850-1873 (SR1). They contain information about who bought and sold a house, the
56 Korevaar, ‘Reach for Yield and Real Estate’; A discussion about housing quality in the context of historical rents in Amsterdam can be found in Korevaar, Eichholtz, and Lindenthal, ‘Long Run Affordability and (Ine)Quality’.
57 Eitrheim and Erlandsen, ‘House Prices in Norway, 1819–1989’.
58 Except for me, Rodney Edvinsson, Klas Eriksson and Emelie Carlsson are also been part of the project.
59 Wikström, ‘Fastighetsforskning i stadsarkivet: en översikt av källmaterialet intill 1850’, 34.
27
size of the plot, if it was a stone house or a wooden house (gård or trägård), price and debt relations between the contractors.60
The city administration was reorganized in 1660 and from thereon records of house transactions can be found in the civil records (civilprotokollen). Land and house transactions are from 1679 registered separately in the so-called minutes of transactions (uppbudsprotokoll).61 Civilprotokollen and minutes of transactions included identical notations about house transaction until the mid-1700s.62
The statistical methods that we have used to create a price index all require that each sale can be connected to an individual house. But before 1720, the city blocks had no official names, even if informal street names and other geographical information were occasionally included. This information can be combined with the size of the plot. With the help of old maps and so-called ground plot books (tomtböcker), the house can be identified. One problem with maps is that they sometimes include planned buildings that never were constructed. Nevertheless, because we start with the sales of existing houses, the risk of identifying a house on the map that never existed is small.63
The geographical information provided by minutes of transactions was initially sparse. All confirmed sales were registered in the protocol of the city’s registry (stadens registratur) and information on the location was often included there. Therefore, those protocols are used as far as 1725.
Sometimes block numbers are not included making individual houses hard to separate. Again, maps can be used to locate the buildings. Another great resource is the numerous registers that have been compiled and digitalized by Stockholm’s city archive through which buildings sometimes are identifiable.64
For the years 1800 to 1839, a register of all sold properties is available, which rests on the minutes of transactions, but is more compactly written and therefore more easily accessible. Since 1800, there is no longer any information on whether the houses sold were made of stone or wood.
60 Before 1600, information about real estate transactions can be obtained from the city’s memory books (tänkeböckerna). An introduction to them can be found in Pettersson, Stockholms stads tankeböcker. Funktionell texthistoria 1476-1626; For a discussion about memory books as a soruce material for house sales, see Franzén and Söderberg, ‘Hus, gårdar och gatubodar. Fastighetspriser i Stockholm och Arboga 1300–1600’; See also Franzén, Sturetidens monetära system, 26ff.
61 1716, 1719 and 1721 are missing.
62 Wikström, ‘Fastighetsforskning i stadsarkivet: en översikt av källmaterialet intill 1850’, 43f.
63 Stockholm’s city archive provides a large selection of digitalized maps and ground plot books on their website. They are also to be found in the archive, see D5 Tomtböcker and Register över handritade kartor.
64 See Register över fastighetsuppbud och upplåtelser 1600-1649 and Fastighetsregister 1675-1875.
28
Table 2: Main archives used in the present thesis.
Period Type of numbers before 1810 must be converted. This has been achieved with the help of the city archives register “Registernyckel för fastigheter 1730 och 1810”.
This register also connects each block with a street name and number.
Combining this information with the comparison of historical street names from old maps, each block can be geographically located.
The database produced does not cover all sales: first of all, non-legal sales are obviously not included. This only applies to cases where both the seller and the buyer had an interest in hiding the transaction from authorities and this number is probably very small. Secondly, many of the sales did not include the whole property but only parts of it. Our sources do not provide information about what it could mean to buy parts of a property, which is why these sales have not been included. Thirdly, houses that have been impossible to identify have been excluded. This problem concerns almost exclusively houses in the poorer, more peripheral, parts of the city.
For each sale, it is also possible to attain information on the name of the buyer and seller and their formal occupational status. This makes it possible to discern any systematic trends in gender, marital status, and class composition among dealers of real estate in early modern Stockholm. From titles, it is also possible to determine if a participant was wellborn or not.
Real estate quality
A major problem when constructing house price indices is to separate price increases from quality changes. In our data, we can separate wooden houses
29
from stone houses between 1730 and 1800. Stone houses yielded, on average, higher prices. So, if someone bought a wooden house and replaced it with a stone house, it might look like there was a price increase when it could actually be a different house that was sold. Since this information was only available for parts of the sample, we have not used this information when constructing the price index in 1730–1875.
Another method is to compare the Old Town with the rest of the city. The Old Town has not changed very much since the seventeenth century and should be less sensitive to quality changes. This is done in both the articles presenting our price indices.
In the article describing the price index for Stockholm 1730–1875, we used information from an archive over building permit applications. Before starting to build new houses or engage in large-scale renovations, the agent had to apply for a building permit. The applications are archived from 1713 to 1875 in the archive of Byggnadsnämndens expedition och stadsarkitektkontor, which includes both planned new buildings and renovations. When we, in the paper A real estate price index for Stockholm 1730–1875, estimated an index with a repeated sales method, the permits were used. Each sales pair for which a building permit application had been handed in between the two sales was removed from the sample.
Another way is to look at the ratio of price and taxation value. The latter is supposed to account for qualitative changes. Tax records are easily accessible from 1856 in the directory of Stockholm (Stockholms adresskalender).65 From 1743 and onwards, a “window tax” was collected (houses were taxed based on how many windows they had). The tax records provide assessment values.66 A price-to-ratio method is employed in one of our papers for which we have used information from Stockholms adresskalender, the address calendar of the city.
Example of a real estate transaction
To understand how the source material may look, a picture of a legal sale together with transcription is shown in Figure 3. This one was written on 14 July 1800 and concerns a house in the northern parts of the inner city.
In Swedish, it reads as follows:
Jungfru Anna Elisabeth Brefving låter uppbjuda, en å Norrmalm vid Norrtullsgatan uti qwarteret Trumpetaren under n. 127 å ofri eller Stadens
65 Stockholms stad, ‘Stockholms adresskalender’.
66 Stockholms magistrat- och rådhusrätt, F 10, Fastighetsvärderingar and
Byggnadsnämndens expedition och stadsarkitektkontor, F 1 AA, Bygglovsritningar 1713–1874.
30
grund belägen fastighet, som aflidne comministern Ludvig Bernhard Ahngers änka Margareta Morström till samma försåldt för 722 Riksdaler 40 skilling 8 runstycken Riksgäldssedlar enligt köpebref af d: 30 Maj 1800.
It reads as follows, in my English translation:
Spinster Anna Elisabeth Brefving hereby announces that she is buying a property on Norrmalm at Norrtullsgatan in the quarter Trumpetaren number 127, on unfree or the City’s land, from the deceased priest Ludvig Bernhard Ahngers widow Margareta Morström, to a price of 722 Riksdaler 40 skilling and 8 runstycken Riksgäld notes, according to the sales letter of 30 May year 1800.
Figure 3: Example of minutes of transaction.
Source: Stockholms magistrat och rådhusrätt, A6a, Lagfarts-, uppbuds- protokoll. Volume 49.
From this protocol, we can obtain the following information:
- Name and marital status of the buyer.
- The exact location of the property.
- That the land of the property was not included, because it was owned by the City.
- Name and marital status of the seller.
- Name and title of the seller’s deceased husband.
- Price.
- Date when the sales contract was signed.
- That neither buyer nor seller was wellborn.
31