The chosen research population was the green patents filed in Brazil. The choice is because green patent application examinations are speedier, i.e., whether or not to grant the patent is made faster than for other patents. In addition, it is a strategic sector worldwide, in times that demand conscious use of natural resources, to allow the development of technological innovations that improve human quality of life while coexisting in harmony and balance with nature.

The main data source for the index was the Global Patent Index (GPI), because it offers the largest number of fields for exporting a large volume of data. The Derwent Innovation Index (DII) was used to collect the citations that patents received from other patents. The third source of information chosen to complement the data is the Brazilian IP office (INPI) database; it provides data on renewals paid by the holders and information relating to the patents’ date of grant.

The search strategy used in the GPI was EVDE (event description) = “green patent”; the search, conducted in November 2019, recovered 631 patent family documents. Patents that did not contain the code “b27b” were removed from the population; this code indicated applications granted entry into the Green Patents Program and patents related to utility model. There remained 478 patent families.

Each of the three indicators is assigned a weight from 1 to 100; the final grade will be the sum of each indicators’ score divided by three.

For this, the following calculations were made:

a) Family size: the number of countries where the patent applied for protection makes up 90% of the indicators’ score; triadic family and PCT application make up 5% each. Patent families were thus divided because the literature on triadic patents – patents that have applied for protection in the main world markets: Europe, Japan and the United States – may be used as an indicator of “patent intensity”, due to the importance of the three offices involved, in addition to being used to compile high-value inventions when comparing countries (Organisation for Economic Co-operation and Development, 2009). Martínez (2011) reinforces this idea by reporting that there are different indicators of internationalization and geography of inventions, based on data from patent families that seek to capture different effects, and triadic family is one of the most used. Researchers and statisticians use this indicator to assess a country’s most valuable patents (Martínez, 2011).

Counting patent families was conducted according to the number of countries in which the depositor sought protection, i.e., geographical scope size (GS). To make a fair calculation, considering the chosen population, patents were leveled by the largest number of components presented by a given invention within the population. In this case, the rule of three is performed; the invention that presented the largest number of protection requests in different countries was BR 112014000125, with 35. Therefore, the equation is:

After calculating the GS, 5 is added for patents with a triadic family and 5 more for those with a PCT request.

b) Renewals: payment of maintenance fees begins 24 months after the application date. Fees increase over the years, so it is necessary to calculate two items: whether the patent is being paid and its current stage of payment (CSP).

To calculate whether the patent is active or when it was active, the number of renewals that the depositor should have paid (PR) is initially ascertained, and then compared with the fees paid, thus patent life percentage (PL) is identified. This is done using the following equation:

The use of the year 2019 to calculate the number of renewals that should have been paid (instead of 2020) is justified by the fact that, in the Brazilian IP office, applicants have three months to pay patent fees in the normal mode and another six months in the extraordinary mode. Due to this payment period and the margin of tolerance as adopted here to avoid injustice, there are cases when depositors who have paid fees on time can exceed the PL by more than 100%; these values will be corrected back to 100%.

The second renewal indicator equation intends to assess the stage of payment. It is necessary to locate the largest number of renewals that a patent within the Brazilian green patent population has, i.e. 16 paid renewals (which are actually 14, due to the first two unpaid years). It should be remembered that an invention patent may remain active for up to 20 years, if its fees are paid on time. Therefore, 100 was divided by 14 (= 7.14), so a patent scores higher for each renewal paid.

After this, for the final equation of the calculation of renewals (RC), we add the two results – PL percentage and CSP – and divide by two:

c) Number of forward citations: First of all, it is important to know how many citations the patent received per year (CPY), taking into account the years in which the invention became publicly accessible, i.e. the year of publication (YP) is taken into account. The equation is the following:

The second step is leveling patents according to the highest-scoring patent from the research population, i.e. the patent with the highest CPY percentage based on the maximum registered value (PCPY-MV). In this case, the highest-scoring patents were BR PI0913013 and BR 112014028704, which reached the maximum CPY score, with 24 citations per year. Therefore, the final equation for citation calculation is:

See Figure 1 for the equations that make up the PVI:

Figure 1 Note: ^*Within the Brazilian green patent population, the highest number of applications for protection in different countries was 35; ^** Within the Brazilian green patent population, the highest number of citations per year an invention received was 24.

The general PVI equation may be consulted in Appendix A: Final PVI (Cativelli, 2022) equation.

To determine how the patent population was distributed in the PVI, the population of 478 patents was divided into four parts. Quartile 1 (Q1) comprises 25.1% (120) of Brazilian green patents – the highest-scoring patents, between 80.3 and 28.6. Quartile 2 (Q2) comprises 24.9% (119) of the patent population, with scores between 28.4 and 21.3. Quartile 3 (Q3) comprises 25.3% (121), with patents scoring from 21.2 to 14. Quartile 4 (Q4) comprises the remaining 24.7% (118) patents, with scores between 13.8 and 0. Table 1 details the quartiles’ descriptive statistics.

Table 1 shows that family and citation indicators are concentrated in Q1, reaching much higher mean and median than other quartiles. In addition, Q1 also stands out regarding data discrepancy, with a high standard deviation, especially in the citation indicator. This may demonstrate the existence of patents that display a much higher volume of the indicators evaluated than other patents in the group. This determinant may also be seen in the maximum and minimum scores each patent received in each quartile, with Q1 scores ranging from 80.3 to 28.6 – the highest span among the four groups.

Table 1

Indicators of patent value	Mean	Median	Standard deviation
Q1 (120)
Family	11.7	12	6
Renewal	6.4	7	2.9
Citation	13.3	5	22.3
Overall Score	40.2	36.9	10.3
Q2 (119)
Family	2.2	0	3
Renewal	4.3	5	1.8
Citation	0.7	0	1.9
Overall Score	24.2	23.8	2.1
Q3 (121)
Family	0.9	0	2.1
Renewal	1.5	1	1.4
Citation	0.2	0	0.8
Overall Score	17.9	17.8	1.7
Q4 (118)
Family	0.4	0	1
Renewal	0.5	0	0.9
Citation	0.1	0	0.4
Overall Score	3.5	0	4.5

Source: Elaborated by the authors (2021).

As expected, numbers significantly plummet at each quartile, and the renewal indicator is the one that guarantees patent scores. In addition, among the indicators assessed, renewal is balanced among quartiles, with the following means: 6.4 (Q1), 4.3 (Q2), 1.5 (Q3) and 0.5 (Q1).

On the other hand, the citation indicator presents the highest score difference among quartiles, with the following means: 13.3 (Q1), 0.7 (Q2), 0.2 (Q3) and 0.1 (Q4). The highest concentration of cited patents is in Q1; the other quartiles display far lowers numbers.

In order to disclose the data behavior that justifies the inclusion of inventions within a given quartile, an analysis of the proportions of the value indicators present in each quartile was conducted. Features related to depositor profile, year of deposit and IPC area were also analyzed.

The first step was to assess the presence of the value indicators – family, PCT, triadic, renewals and citation – in each quartile. This may initially sound biased, as it is already expected that Q1, the highest-scoring group, will present higher numbers. However, this is not an assessment of quantity, but of whether or not the indicator is present; in the case of the renewal indicator, it will be also ascertained whether the patent remains active (whether holders are keeping up with payments). Results may be seen in Figure 2 below.

Figure 2 Source: Elaborated by the authors (2021).

As expected, Q1 (120) scores higher across all value indicators. Almost all the patents in this group were shown to have PCT (111) and family (116), which indicators are usually seen in conjunction, as it will be more advantageous for holders to file a PCT application when they wish protect their patents in several countries.

The same phenomenon occurs in Q2 (119), Q3 (121) and Q4 (118): Q2 has 52 patents with PCT application and 58 with family; Q3 shows 22 patents with PCT application and 29 with family; Q4 has 16 patents with PCT application and 19 with family.

It also calls attention that, in Q1, there is a large number of patents with citations: 106; this means only 14 patents in this group are not cited. This indicator drops drastically in the other groups: Q2 has 27 cited patents, Q3 has 13 and Q4 has 10.

As previously mentioned, the renewal indicator shows the most balanced scores among quartiles; however, here it was assessed whether renewal payments are up-to-date. None of the patent renewals in Q4 were up-to-date.

Therefore, patents were found to have a somewhat predictable behavior according to the quartile to which they belong: the better positioned in the PVI, the higher their chances of presenting the indicators here assessed. And, as proved in Q1, the major finding of the present research is that the evaluated indicators go together and relate, i.e. these value indicators are not scattered among quartiles; their behavior is such that, when one indicator is present, chances of the others being present as well are higher. The logical justification for this is that, if an invention has broad market potential, depositors will also protect the patent outside their own country; the most practical way of doing so is filing a PCT application. Thus, the higher the number of countries where a patent is published and translated, the more likely it is to be cited. And if patents prove profitable for their holders by then, they will continue to pay for renewals on time.

The next features to be assess are depositors’ profiles – whether or not they are foreign – and to what type of organization they belong. When ownership was mixed, only the first holder was taken into account. In addition, it was verified whether inventions were granted in Brazil. These numbers are shown in Figure 3.

Figure 3 Source: Elaborated by the authors (2021).

The most prominent feature in Figure 3 is the high number of patents from foreign companies in Q1: 106 companies, 98 of which are foreign. Q2 also presents significant numbers, with 57 patents belonging to companies, 29 foreign ones. In Q3, there is a balance between patents belonging to companies (43) and natural persons (51); only 8 foreign holders are in this quartile. In Q4, the number of natural-person depositors is higher (65); only one is foreign.

About patent grants, Q1 and Q2 are practically identical, with 90 and 89, respectively; in Q3, 54 patents were granted; in Q4, only 17.

It is also possible to notice the following behavior: highest-scoring quartiles have the most expressive numbers of foreign company holders with granted patent, whereas the lowest-scoring ones show higher concentrations of natural person holders, who are Brazilian and have less chances of being granted the patent.

Data on the year of filing a patent application is thought relevant for assessing whether the most recent patents are in the lower quartiles. This is because newer patents are at a disadvantage regarding citations and grant, in addition to other aspects to be addressed later. Results may be seen in Figure 4.

Figure 4 Source: Elaborated by the authors (2021).

It is observed that Q1 has the oldest patents; it peaks in the 2010, with 21 records, as indicated by the blue line. Another curious fact is that Q1 has more patents filed before the Green Patents Program (PPV) was launched (2012) than after: 68 patents from 2004 to 2011 and 52 from 2012 to 2018.

In the other quartiles, the number of patents filed before the PPV is low, and decreases according to the quartile to which they belong: Q2 has 23 patent filings, Q3 has 13 and Q4 has 5.

The data in Figure 4 clearly shows that the PVI’s highest-scoring patents are concentrated in the years before 2014, as the blue and orange lines – representing Q1 and Q2, respectively – show significant numbers in this period.

About Q3 and Q4, the opposite occurs: they begin to register more deposits as of 2013. Two explanations may be found for this phenomenon. The first and most obvious is that the older a patent is, the more chances it has to prove its value, as it will be more cited, in addition to patent holders already having a more defined marketing strategy, and, if necessary, having already applied for protection in the countries of their possible competitors.

The second explanation is that, during the last decades, the Brazilian IP office and technological innovation centers have promoted and encouraged applications, stimulating inventors to patent their inventions.

According to the Organization for Economic Cooperation and Development (Organisation for Economic Co-operation and Development, 2011), there was a drop in patent quality by approximately 20% between 1990 and 2000 in almost all countries studied. The reason for this is that the rush to protect small improvements in products or services may burden patent offices, which causes true innovations to take longer to reach the market, and reduces the potential for revolutionary inventions.

Regarding the overall ranking, Brazil in not among the ten highest-scoring patents. This list may comprise patents of international owners of green inventions protected in Brazil. Table 2 shows how scores were achieved.

Table 2

Patent Number	PCT	Triadic	Family (number)	Family (%)	Renewals (number)	Renewals (%)	Citations (number)	Citations (%)	Total
BR PI0913013	Yes	No	21	59.00	9	82.0	48	100.00	80.3
BR PI0718314	Yes	Yes	17	53.70	9	77.0	164	97.62	76.1
BR 112014028704	Yes	Yes	14	46.00	5	68.0	96	100.00	71.3
BR PI0913059	Yes	No	16	46.10	9	82.0	46	63.89	64.0
BR 112014014000	Yes	Yes	15	48.60	5	68.0	36	75.00	63.9
BR PI0911588	Yes	No	17	48.70	9	82.0	53	55.21	62.0
BR PI0916189	Yes	No	20	56.40	9	82.0	28	38.89	59.1
BR PI10085655	Yes	Yes	24	71.70	8	78.5	23	19.17	56.5
BR PI1013007	Yes	Yes	26	76.90	8	78.5	16	13.33	56.2
BR PI0820556	Yes	Yes	12	40.90	10	86.0	57	39.58	55.5

Source: Elaborated by the authors (2021).

The ten best-evaluated patents have PCT application, six of them have triadic families and the invention was protected in 12 to 26 different countries. Regarding paid renewals and their percentage, it is also necessary to see Table 3 below; these patents were filed between 2007 and 2013, and only one is inactive. These inventions received 16 to 164 citations over the years; the lowest citation record per year (counted from the date of publication) was that of patent BR PI1013007 with 3.2, and the largest were those of patents BR PI0913013 and BR 112014028704, with 24 citations per year.

As for other features of these patents, two were not granted in Brazil; the holders are all foreign, with a predominance of companies from the United States, especially Xyleco, with inventions involving biomass processing. Further details on patent features may be seen in Table 3 below.

Table 3

Patent Number	Application year	Active patent	Grant	Applicant country	Applicant name	Patent title
BR PI0913013	2009	Yes	Yes	United States	Xyleco	Processing Biomass
BR PI0718314	2007	No	No	United States	Medoff Marshall	Processing Biomass
BR 112014028704	2013	Yes	Yes	United Kingdom	Isis Innovation Limited	Optoelectronic devices with organometal perovskites with mixed anions
BR PI0913059	2009	Yes	No	United States	Xyleco	Method of fermenting low-molecular weight sugar to ethanol
BR 112014014000	2012	Yes	Yes	United States	Monsanto Technology	Plant growth-promoting microbes and uses therefor
BR PI0911588	2009	Yes	Yes	United States	Xyleco	Method for obtaining animal feed by processing of biomass
BR PI0916189	2009	Yes	Yes	United States	Xyleco	Lignocellulose biomass material fermentation method
BR PI1008565	2010	Yes	Yes	United States	Xyleco	Method for reducing the recalcitrance of a biomass feedstock and a method for producing a product
BR PI1013007	2010	Yes	Yes	United States	Xyleco	Processing Biomass
BR PI0820556	2008	Yes	Yes	New Zealand	Lanzatech New Zealand	Novel bacteria and methods of use thereof

Source: Elaborated by the authors (2021).

As Table 3 shows, six patents belong to Xyleco – the holder with the highest number of green patent applications of the research population, totaling 24. It is a private scientific research and manufacturing company that develops processes to convert biomass into useful products (Xyleco, 2020). Xyleco may have seen Brazil as a true competitor in this industrial branch, as its immense national territory, almost entirely located in tropical and rainy regions, offers excellent conditions for large-scale biomass production and energy use. In addition to the production of alcohol, burning in furnaces and boilers, and other non-commercial uses, biomass has great potential in the electricity sector (Agência Nacional de Energia Elétrica, 2005).

In addition, the patent that appears in the GPI database as owned by Marshall Medoff (PI0718314), in the INPI database is registered by Xyleco: in fact, there are seven patents belonging to this company, as they were invented by the company’s founder – Marshall Medoff himself.

Patent BR PI0718314 – which holds the second highest PVI score – was not granted in Brazil nor have its renewals been kept up-to-date, therefore, it may be used in Brazil for free. In this sense, the PVI shows high usability.

The story of Marshall Medoff is fascinating; he has no scientific background, and spent15 years recluse in a garage trying to find answers to his questions; he believes people were failing because they were trying to overcome nature, instead of working with it (Stahl, 2019). His breakthrough was being aware that there is a lot of energy in plant life, stored in the form of sugar molecules that, once accessed, can be converted into transport fuel. This sugar is almost impossible to extract inexpensively and cleanly, as it is firmly trapped inside the plant pulp. It should be noted that cellulose is rich in sugar and is the most abundant biological material on the planet (Stahl, 2019). With the visibility of his discoveries, Medoff began to attract investors and hire scientists to help him. He is even believed to be able to expand enough as to compete with the oil industry (Stahl, 2019).

The company that ranks third is from the United Kingdom: Isis Innovation Limited, a private technology and research company from the University of Oxford; in 2016, it was renamed Oxford University Innovation (Oxford University Innovation, 2020). It procures the university’s advantage, by working with staff members and students to apply their knowledge and research; its vision is to create a worldwide leading innovation ecosystem, having the University of Oxford at its center (Oxford University Innovation, 2020).

The United-Statesian Monsanto Technology, which has Monsanto as its parent organization, ranks fifth. Monsanto provides agricultural products including special seeds, agricultural chemicals, plant varieties and agricultural productivity solutions to farmers worldwide (Monsanto, 2020). The company does not have a good reputation in terms of sustainability: it is a world leader in producing the herbicide glyphosate and genetically modified seeds.

The lesson here is that even if a given institution develops green patents, that does not necessarily mean it is concerned about the environment. Many companies may be seeking to improve their image by featuring sustainable patents in their portfolios; this practice is known as greenwashing. As Paviani (2019) explains, environmental protection gains space in business policies, especially, to gain credibility with consumers, but the extent to which green marketing agrees with the effective practices of corporations should be called into question.

Closing the top ten, one New Zealander holder is Lanzatech New Zealand. It claims to be turning the global carbon crisis into a raw material opportunity, by recycling carbon from industrial gases, synthesizing gas from any biomass resource and reformed biogas (Lanzatech, 2020).