What can we learn about patents and innovation from the past?

The history of industrial revolutions provides valuable insights into how patents and patenting systems influence innovation. It also reveals the extent to which inventors have relied on other ways to protect their ideas, including secrecy, sharing and showcasing them at international exhibitions.

We are now firmly in the middle of the fourth industrial revolution, according to the World Economic Forum. This is a period of rapid and widespread technological change characterised by the development of cyber-based and digital innovations (World Intellectual Property Office, WIPO, 2023).

The wave of innovation is accompanied by a global increase – for the fourth year in a row – in the number of patent applications filed (see Figure 1).

By their very nature, we know that patents and innovation are related to one another. Patents are a property right given to an inventor, allowing them the sole legal right to use, sell or license their invention in exchange for disclosing the new knowledge of their invention to the public.

This relationship is not fully understood, but we know that it is not straightforward, not least because of measurement errors in determining the true level of innovation in an economy – not every patent is an invention; and not every invention is patented (Schmookler, 1966).

Figure 1: Global patent applications

Source: WIPO

Despite uncertainty around the exact role that patents play, it is clear that technological development is key to achieving sustainable economic growth over the long run. Further, identifying factors that provide incentives for innovation is a key priority for governments, institutions and businesses across the globe.

The current period of innovation is not unprecedented. It follows previous instances of rapid technological development that provide useful lessons for understanding the relationship between an incentive system like the patent system and the level and direction of innovation in an economy.

A series of revolutions

The first industrial revolution, which originated in Britain during the second half of the 18^th century, coincided with a sharp increase in patenting activity as the country ‘became more inventive’ (Sullivan, 1990).

The second industrial revolution saw the rise of the United States as a major technological leader during the late 19^th century. This revolution centred on mass production and general-purpose technologies such as electricity. It too was accompanied by a rapid increase in both patenting and global reform of patent systems to follow the US archetypal system (Khan, 2005).

What has been called the third revolution came in the last few decades of the 20^th century with waves of information technology (IT) and computing innovations that enabled the automation of production and processes. These were embedded at every level of society (Nuvolari, 2019).

The fourth or digital revolution comes then with similar expectations of transformation: technological disruption and economic opportunities that have not previously been possible, accompanied by an increase in patenting activity.

Despite the obvious differences in temporal and geographical scope, are there lessons from history that are relevant for our current industrial revolution?

The short answer is yes, and it relies on the similarities between these revolutions. They all took place alongside the development of institutions that were designed to foster innovative activity, to drive new inventions and to appropriate the returns from them.

The longer answer – outlined below – reveals that the extent to which patent systems, historical or contemporary, actually matter for innovation, and crucially, how they matter is strongly debated.

Simply put, we do not yet fully understand this complex relationship, but almost two centuries of economic history provide a much-needed evidence base.

What determines patent levels in industrial revolutions?

The first industrial revolution

England, and later Britain, has one of the oldest patent systems in the world, with legislation dating back to 1624. Very few patents were granted until the 1750s (an average of seven a year) largely because of the system’s notorious complexity (Macleod, 2002).

After 1750, the number of patents granted began to increase, roughly coinciding with the early years of the first industrial revolution (1760-1830) to an average of 500 a year by the 1840s, as Figure 2 shows (Sullivan, 1989).

The patent system in Britain was not like today’s. It was expensive and time-consuming to obtain a patent, with applicants having to apply through multiple offices across the country in a process that could often take weeks or months (Bottomley, 2014).

The high fees – around more than ten times a labourer’s average annual wage – influenced patenting in two ways. First, they restricted access to the system to those with the means to pay (Macleod et al, 2003; Billington, 2021).

Second, they may also have acted as a filter or screening mechanism for stopping low-quality or trivial patents, thereby encouraging inventors to patent because there were fewer potential legal challenges in the patent pool (Bottomley, 2014).

These characteristics muted the level of patenting activity. Britain’s population was 22 million in 1850 (Broadberry et al, 2015), yet only a few hundred patents were being granted. Clearly, then, the pricing structure is important, especially when considering the role that price played in another patent system, that of the United States.

Figure 2: Patents in England

Source: UK Intellectual Property Office

The second industrial revolution

A second, more rapid period of technological change, dominated by the United States, began in the late 19^th century and ended before the First World War. In this period, the United States overtook Britain in terms of industrial production and economic growth (Crafts, 2012). There was also an even more pronounced increase in the usage of the patent system (in both countries) to protect inventions (see Figure 3).

The US patent system had matured by the time of the second industrial revolution, having been adopted in 1790 and modified in 1836, becoming the de facto model patent system. It was, however, fundamentally different to the British system in both cost and structure.

US patents cost considerably less than those in Britain – around $30 per patent or £6 in 1860 (see Table 1). They were administered centrally through one office and were explicitly examined for novelty.

So, in a country where innovation was increasing rapidly and driving the second industrial revolution and its associated economic growth (Khan and Sokoloff, 1988), the patent system was cheaper and easier to access.

But the patent system did require a higher burden of proof on the part of the inventor. As a result, it was an industrial revolution accompanied by an entirely different set of incentives and barriers for inventors looking to capture economic returns to their innovative pursuits.

Table 1: Average cost of a patent in GBP

	1850	1870
Great Britain	£375	£175
United States	£13	£6

Source: Hancock, 1850; Khan, 2008

Indeed, the first substantial patent reform in Britain, which occurred in 1852, was a shift towards a US-style system. The change reduced the number of offices to one and cut the costs for patent protection. As a result, the number of patent applications rose into the thousands.

This also raises a key issue with patent systems – the effect of policy on access to the system. Clearly, Britain did not become doubly more inventive in the year between 1852 and 1853. Cheaper patents mean more patents, but not necessarily better ones (Nicholas, 2011).

Alternatives to patenting

A brief examination of raw patent data seems to suggest that both the US and British patent systems, while different, were capable of allowing large numbers of inventors to patent their inventions.

But they did so with different screening mechanisms for stopping useless or spurious inventions. Britain’s system favoured higher fees, while the United States focused on examining quality and novelty itself.

One might conclude then that a system with low fees or examination, or both, might encourage sufficient levels of innovation to support an industrial revolution. But there is a caveat – and a large one for historians and policy-makers alike – that is the driving force behind much of the patent research in economic history and related fields over the last two decades.

This caveat is that invention does not occur solely within the patent system, so any attempts to provide incentives for innovation solely through patent systems are likely to overlook a considerable degree of inventive activity, and thus growth opportunities. Patent policies can most readily deal with the fact that not all patents are inventions; but the fact that not all inventions are patented is much trickier to resolve.

History again provides evidence to address invention outside the patent system. One key source in this regard is the set of publications from World’s Fairs and Exhibitions, which were frequently held throughout the second half of the 19th century – the 1851 Great Exhibition being the first.

These were venues to showcase cutting-edge innovations from across the world. For example, of the nearly 14,000 innovations presented in the Crystal Palace in 1851, just 12% were patented – with many more patents going to mechanical inventions than to chemical ones (Moser, 2005).

The likelihood of patenting when considering this additional measure changes based on the technology type, which is referred to as the propensity to patent. So, industries and technologies with a higher propensity to patent are likely to respond more readily to innovation policy aimed at patents by virtue of the nature of the inventions in those industries.

It is much easier to reverse-engineer machinery, for example, than a chemical process. Because of this, there are important differences in the need to use patent systems to protect innovations depending on the type of innovation.

Figure 3: Patenting and the Second Industrial Revolution

Source: US Patent and Trademark Office; UK Intellectual Property Office

Inventors in industries with a low propensity to patent – such as brewing, chemicals, earthenware and food processing – usually opted to conceal their inventions and keep them secret instead (Moser, 2012; Nuvolari and Sumner, 2013; Lane, 2019).

The key advantage of this strategy is that it doesn’t carry the costs of taking out a patent, or the sharing of valuable knowledge with competitors through the publication of the patent specification.

Such innovations were also granted a degree of protection by virtue of the inability to reverse-engineer, say, a glaze on pottery once it has been fired. Imitation was through experimentation, which was costly.

Industrialisation also occurred in countries that did not possess patent systems, namely Switzerland and the Netherlands. Switzerland would not adopt a patent system until 1888, while the Netherlands abolished theirs in 1869, before reinstituting it in 1912 (Schiff, 1971).

An examination of the inventions submitted at World’s Fairs by nationals from these two countries reveals the direction that innovation took in the absence of the use of patents. Swiss inventors often presented chemical innovations, while Dutch inventors presented more food processing innovations, particularly those relating to margarine (Moser, 2005). The firm that would go on to become Unilever emerged during this patent-less period as a margarine producer.

The evidence demonstrates that in the absence of a patent system, innovation is directed towards those technologies that are better protected through secrecy.

Alternatively, some inventors openly shared their inventions with competitors, in a practice known as collective invention. This happened regularly in metallurgy during both the first and second industrial revolutions (Allen, 1983).

The practice was cooperative, typically taking place with innovations that were not patentable, such as in the blast furnace industry in the north of England during the first industrial revolution where optimal designs were shared and published. This was on the understanding that future knowledge and innovation would also be shared.

The development of the steam engine, an important invention for the first industrial revolution, was also driven by collective invention (Nuvolari, 2004). Clearly, patents have little effect in this type of cooperative innovation.

Conclusion

In 1958, the economist Fritz Machlup conducted an economic review of the benefits of a patent system on behalf of the US Senate. His conclusion succinctly summarised the prevailing thought on the role of patents: ‘If we did not have a patent system, it would be irresponsible, on the basis of our present knowledge of our economic consequences, to recommend instituting one. But since we have had a patent system for a long time, it would be irresponsible, on the basis of our present knowledge, to recommend abolishing it’ (Machlup, 1958).

Almost 70 years later, Machlup’s conclusion still resonates, but research in economic history has advanced our understanding of patents much further. History reveals the extent to which inventors relied on other methods of protection, whether that was secrecy, sharing or dissemination at exhibitions.

History also reveals how the design of patent systems can influence what ends up being patented – and that patents can direct innovative activity. Although we are not yet at a satisfactory conclusion on patents, history is helping to drive our understanding forward.