Patent protection can make an invention commercially viable, but also limits its use by others. Explore two historical applications and decide whether protection is needed, and on what terms.
Based on research by Homer Wagenaar and Christopher L. Colvin, published in The Economic History Review (2026).
Created in 1815 after the Napoleonic Wars, the United Kingdom of the Netherlands joined the former Dutch Republic in the north with the former Austrian Netherlands in the south. A single patent system followed in 1817, administered from the capitals, Brussels and The Hague.
During 1817–1830, the northern provinces had a patent system but had not yet undergone a broad industrial transformation. Industrialisation gathered pace later in the nineteenth century and continued after the Netherlands abolished patents in 1869.
The southern provinces industrialised earlier, despite sharing the north’s patent law and administration. They were undergoing the transformation that made Belgium, independent after 1830, the world’s second industrial nation. Both new inventions and technologies introduced from abroad mattered to this process.
Joel Mokyr (1976) drew on aggregate patent statistics for the Low Countries to illuminate this contrast: north and south shared patent institutions, yet followed different paths of industrialisation.
The contrast poses the paper’s central question: what can a patent system achieve when the same institutions serve regions with different economic conditions? Having a patent law was not enough, on its own, to bring about industrialisation.
Where were people and industrial employment concentrated? Switch between the population and industry views.
Borders and province names
The 19 provinces used in the paper’s geographical analysis.
Blue lines show the later international borders following the Treaty of London in 1839, not borders between separate states in 1819.
The 1817 system protected original inventions and the introduction of foreign technologies. Officials could reject an application, reduce its fee, shorten its term, or attach conditions. Their task was to weigh novelty, usefulness, and the applicant’s financial means.
“to reconcile general interest with individual interest, that this enjoyment be limited to the shortest possible term”
‘Report to the King’, 17 June 1817, no. 1172, Hartsinck & Blanckenay file, NL-HaNA, 2.04.01, 4040. Translation by Wagenaar and Colvin. The report argues for sufficient reward while limiting the public burden of exclusive rights.
We collected and digitised the complete set of patent applications submitted in the United Kingdom of the Netherlands during 1817–1830 from the original archival files, including rejected and withdrawn applications as well as those that received patents. This allows us to examine who applied, which applications succeeded, and how officials varied fees, duration, and conditions.
An application could pass through several assessments of its legal requirements, novelty, and usefulness before officials settled its terms and fees. Patents also formed part of a wider system of support, including subsidies and loans.
Despite high standard fees, applicants came from across social classes. For inventions, medium- and lower-skilled applicants were more likely to receive fee reductions than factory owners, nobility, and landholders.
Technically skilled applicants were more closely associated with patents of invention; commercially oriented applicants with patents of importation. Administrators treated the two patent types differently, consistent with encouraging useful technology transfer as well as domestic invention.
Which provinces had high application rates? Were they also prominent in grants?
These maps show applications and grants per person across the kingdom during 1817–1830.
Patent of invention applications, per capita.
The two maps have different legend breaks; compare the legends as well as the colours.
Officials used discretion to screen applications and adjust protection. The model in the article lets you explore those decisions through two documented inventions. Choose a case and adjust its patent terms: can you make the project happen while limiting the cost of exclusive rights?
Use Reset to return to the starting settings between exercises.
Deciding under uncertainty. Taking out a patent was a risky investment: only a minority proved commercially valuable. Applicants and officials had to estimate an invention’s prospects. The figures below are illustrative, allowing you to explore how those estimates affect the decision.
Estimated earnings before implementation costs, without exclusive rights. This is an assumption, not a recorded return. Zero assumes no earnings without protection; a return covering implementation costs means the project can proceed without a patent. Adjust it to explore your judgement.
Illustrative starting offer—adjust the terms, then compare with the recorded decision below.
Assigned terms were usually 5, 10, or 15 years; patents of importation based on foreign rights could follow their remaining term. Half-year steps here let you explore the minimum protection needed.
Post-grant extensions could be requested, but are not modelled here.
Conditions such as requiring licences for other producers can limit exclusivity. This slider represents their effect on protection; it is not a historical percentage measure. Bent’s 18-month deadline for putting the technology into operation is a separate historical condition, not a percentage on this scale.
The fee follows the historical guilder schedule and could be reduced or waived entirely. Half-year duration steps and the restriction scale simplify the choices.
Historical terms from Wagenaar and Colvin’s replication data. The returns and costs used in the exercise remain illustrative.
You estimate the applicant’s gross return without a patent, then choose whether to grant protection and on what terms. The starting offer and financial values are illustrative; the requested duration comes from the historical record.
The applicant takes up your offer only if it covers implementation costs and the fee, and is at least as attractive as proceeding without a patent. If earnings without protection already cover implementation costs, the project can proceed even when you reject the application. Equality is sufficient.
Effective protection = duration × (1 − restriction)
Private return = return without a patent + 60 × protection
Public cost = 20 × protection
The assessed public outcome includes the benefits of projects that proceed, subtracts monopoly costs and an examination cost of 50 per application, and assigns fee revenue a public value of 0.2 per guilder. We show the difference from a no-patent decision on the same application. Examination costs cancel in that comparison. Private returns and administrative assessments are different concepts; this is not a calculation of realised social welfare.
Reading the chart. Pink shows insufficient protection; the green boundary marks the minimum needed at the chosen fee; amber shows additional protection that adds public costs without inducing another project. The boundary is conditional on your assumptions, not a historical optimum. If the project would proceed without a patent, no protection is needed, even if the applicant would accept it.
Public costs. Exclusive rights are assumed to cost others 20 guilders per effective year of protection. The reported excess cost is the protection beyond the minimum multiplied by 20. These costs arise only if the applicant takes up the patent. They are distinct from patent fees and administrative costs.
Finding sufficient protection. The model in the article connects protection, incentives, and administrative decisions. Here, “Find sufficient duration” finds the shortest sufficient half-year term, lowering the fee when its ceiling requires it. It cannot exceed the recorded request. Rejecting the application removes the fee and protection; granting again restores your chosen terms.
Historical fees. Standard fees were 150, 300, and 600 guilders for 5, 10, and 15 years. Officials could reduce or waive these fees. The fee ceiling follows the granted duration: 30 guilders per year for the first ten years, plus 60 per additional year up to fifteen. Shortening the term automatically lowers any fee above its new ceiling. The requested term limits the maximum duration. Five-guilder fee steps, half-year duration steps, and the restriction index allow exploration of the model; they do not reconstruct every historical policy choice.
You have explored two applications: one for an invention and one for an importation. What happens when officials must decide many, with different costs and prospects? A single offer can give some applicants more protection than they need while leaving others unable to proceed. UKNL officials tailored terms to these differences; the model in the article reflects that practice.
Individual assessment depends on judging applicants’ needs correctly. Gathering information costs time and money; mistakes, favouritism, and abuse can undermine its benefits. Yet the UKNL research finds no evidence of systematic discrimination, and fee reductions widened access for applicants with limited means.
Compare the two approaches using four fixed illustrative applicants, independent of your settings above. Here, officials assess each project correctly. The slider tests how much extra administration that advantage can bear; errors and abuse are not simulated.
Every applicant is offered 10 years of protection, a 200-guilder fee, and the same conditions.
Fees are waived and protection is limited to the shortest term that gets each dependent project started. The importer who would proceed anyway receives no patent.
How much more does individual assessment cost than applying one rule? A 40% increase means administration costs 1.4 times as much. This is an assumption for the exercise, not a historical estimate.
| Application | One rule | Individual assessment |
|---|
Both approaches use the model from section 03 with four fixed hypothetical applicants, independent of your earlier settings. It checks which projects proceed and weighs their additional benefits against exclusivity and administration costs. The common offer restricts exclusivity by 20%. Individual assessment removes those restrictions, waives fees, and rounds the minimum sufficient term up to the next half-year. Costs and benefits are expressed as illustrative amounts in guilders.
Standard administration is assumed to cost 50 guilders per application, or 200 guilders for these four applications. The slider adds a percentage of that baseline: 40% adds 80 guilders in total. It represents additional officials’ time, correspondence, and technical advice; losses from inconsistent decisions are not included. The baseline and percentage range are illustrative.
In the model in the paper, the benefit of tailoring must outweigh its additional costs. This example assumes officials know each applicant’s costs and alternatives; real officials had incomplete information. It compares two specified approaches, not the best possible uniform rule with the best possible individual decisions.
The research offers questions for contemporary innovation policy. These are implications drawn from the historical evidence and conceptual framework, not estimates of the effects of changing modern patent law.
The UKNL system recognised both domestic invention and the introduction of foreign technologies. For policymakers, the question of how to make the national economy prosper extends beyond generating new ideas to helping firms adopt and adapt useful knowledge from fellow innovators and abroad.
Protection can encourage a project while imposing costs on others. The aim is to provide a sufficient incentive, taking the fee into account, without granting unnecessary exclusivity. Activity that would occur anyway needs a different justification for support.
The UKNL system was accessible across social groups despite high standard fees, with fee reductions reflecting applicant characteristics. Policy assessment should examine actual access, costs, and implementation, as well as formal eligibility.
The risk of abuse is a reason to scrutinise discretion, not automatically eliminate it. Safeguards against favouritism should preserve the flexibility to support applicants with different needs.
Acemoglu and Robinson (2012) emphasise economic institutions that give people broad opportunities to participate and innovate. The UKNL patent system offers a concrete example: access extended across social groups, and officials could reduce fees for applicants with limited means. Individual assessment could help widen access when a uniform charge would have excluded useful inventions.
North, Wallis, and Weingast (2009) argue that broad access to economic opportunities depends on impersonal rights rather than personal privilege. Our findings also highlight the importance of scale: discretion was manageable with relatively few applications, while rising volumes in Belgium encouraged more formal procedures. The challenge is to preserve access and responsiveness as the system grows.
The same inherited patent law followed different paths after the partition of the UKNL in 1830. Belgium saw growing use and codification, culminating in reform in 1854; in the Netherlands, neglect was followed by abolition in 1869. Legal rules alone cannot explain the performance of an innovation system. Wagenaar (2026) traces this divergence in TSEG.
This brings us back to Mokyr’s (1976) comparison: shared patent institutions did not produce the same industrial trajectory. Their effects depended on the demand for protection and how the system was administered and enforced.
Wagenaar, Homer, and Christopher L. Colvin (2026). ‘Patently peculiar: Patents and innovation in the United Kingdom of the Netherlands’. The Economic History Review, 1–43. https://doi.org/10.1111/ehr.70163.
Wagenaar, Homer, and Christopher L. Colvin (2026). Supplementary materials for ‘Patently peculiar: Patents and innovation in the United Kingdom of the Netherlands’. The Economic History Review. Supplementary materials (PDF).
Wagenaar, Homer, and Christopher L. Colvin (2026). Data and replication package for ‘Patently peculiar: Patents and innovation in the United Kingdom of the Netherlands’. CERN Data Centre’s Zenodo digital repository. https://doi.org/10.5281/zenodo.23010767.
Wagenaar, Homer (2026). ‘The patent system of the Netherlands in a Belgian mirror, 1817–1869’. TSEG – The Low Countries Journal of Social and Economic History, 23(2), 5–36. https://doi.org/10.52024/mhh4bp30.
Acemoglu, Daron, and James A. Robinson (2012). Why nations fail: The origins of power, prosperity, and poverty. New York: Crown Publishers.
Mokyr, Joel (1976). Industrialization in the Low Countries, 1795–1850. New Haven: Yale University Press.
North, Douglass C., John Joseph Wallis, and Barry R. Weingast (2009). Violence and social orders: A conceptual framework for interpreting recorded human history. Cambridge: Cambridge University Press.
Philips, R. C. M., and E. Buyst (2020). ‘Regional industrialization in Belgium and the Netherlands’. In B. van Leeuwen, R. C. M. Philips, and E. Buyst (eds.), An economic history of regional industrialization, chapter 3. London: Routledge.
.jpg){kind=link}