PHARMARANK
PharmaRank is a pilot project created by DELab UW in cooperation with PZPPF, which enables the creation of economic rankings for companies in the pharmaceutical sector.
The PharmaRank tool allows you to compare Polish producers of pharmaceutical products according to their importance to the national economy using a classification algorithm.
PHARMARANK FUNCTIONS
PharmaRank operates as an interactive application that allows:
- conducting real-time simulations for various indicators
- adjusting the weight values for individual indicators
- making flexible comparisons of individual companies
PharmaRank has features that allow you to:
- generating tables based on selected criteria and user-defined sorting – e.g. comparing top companies by importance for the Polish labor market, or generating tables showing companies that are the most similar in terms of net exports
- classifying pharmaceutical companies according to the adopted indicators – the result is the division of companies into five segments (AE) in accordance with the proposal put forward by the government administration
- creating interactive graphs of changes in individual indicators over time
- a preliminary visualization of the algorithm dividing companies into segments
CATEGORIES IN PHARMARANK
The categories were selected based on macroeconomic analyses of the impact of the pharmaceutical sector on the economy by DELab UW. Additionally, the research results were compared with a microeconomic analysis of companies belonging to the sector. The selection of categories was based on research on the impact of the pharmaceutical sector on the economy.
The choice of impact categories can be changed and depends on the availability of data at the enterprise level. The application allows for the introduction of a wider or different set of categories. In this way, both the ranking results and the assignment of companies to categories (from A to E) change. By default, the tool includes five impact categories (sales, investments, taxes, salaries, net exports).
To meet the criterion of basic comparability, data comes primarily from companies' balance sheets. Their source is the combined databases of business intelligence agencies Biznode and InfoCredit. In the case of missing data, PZPPF member companies had the opportunity to supplement the database in order to improve the comparability of individual categories. The data update process may be continued depending on the needs.
The selection of a category additionally meets the following conditions:
- simple definition – the category should be simple and possible for companies to calculate, among other things, in order to avoid additional transaction costs,
- comprehensibility – the category should be comprehensible to both businesses and public administration,
- auditability – the category should be easily verifiable.
METHODOLOGY FOR SELECTING CATEGORIES AND WEIGHTS
CUSTOMIZING WEIGHTS FOR SELECTED CATEGORIES
The application has a wider set of categories available, which can be selected using the “Select categories for analysis” button.
For example, if we assign weights of 1, 2, 1, 3, 4, the result will indicate the relative importance of each category. The second category is twice as important as the first, the fourth is 3 times more important than the first, and the first and third are equally important.
The results are normalized by relating the values for individual firms to the maximum value in the sector and taking the square root of any negative values (or missing data) by zero.
The basic indicator assumes a value for individual categories from 0 to 100%. If the maximum value in a given category is 10 thousand and the result of the company we are interested in is 2,5 thousand, then we get a result of 50% (the square root of 2,5 thousand divided by the square root of 10 thousand).
HOW DO WE GET THE FINAL RESULT?
To obtain the final result, the values of individual categories are multiplied by the weight for a given indicator and then divided by the sum of the indicator weights (in our example: 11). As a result, the maximum result can be 100%.
The classification of companies is based on the formula:
where: w_n: weight of the nth indicator, k – number of indicators, i_n – value of the indicator of the nth company, i_n max – maximum value of the indicator achieved in the sector (in all years).
Assignment to categories is based on the weights entered by the user – by default, this is a minimum of 40% for category A, 20% for B, 10% for C, 5% for D. Companies below 5% receive category E. In the application, you can select data for the years 2012-2016, as well as – in a separate tab – the 5-year and 3-year (2014-2016) average for individual companies.
METHODOLOGY FOR SELECTING CATEGORIES AND WEIGHTS
Macroeconomic background
The pharmaceutical sector is characterized by high expenditure on research and development activities in the Polish economy. The total amount of intangible assets and tangible fixed assets increased in all analyzed periods. In terms of amount, expenditure on research and development and investment is dominated by large companies. In percentage terms of revenues, large companies with Polish capital stand out, which also increased the value of tangible fixed assets and intangible assets the most.
The sector affects GDP through investment, exports and consumption. The magnitude of this response is presented in the table.
Interpretation: An increase in investment expenditure (total at purchase prices) by PLN 1 will result in an increase in GDP by PLN 0,6117. Of this, taxes on products will increase by PLN 0,017, costs related to employment by PLN 0,2065, etc. An increase in investment expenditure by PLN 1 million will result in an increase in the number of employed persons by 4,83 persons.
Main channels of influence
Assessing the impact on the economy at the enterprise level requires a different methodology than that used by the macroeconomic approach. Using the current reporting of capital companies, it is possible to analyze only direct effects more precisely, because the study of indirect and income effects is feasible within the framework of a network analysis of the entire economic ecosystem.
For the purposes of PharmaRank, the following pro-development categories were selected: export, investment, sales (simulated by consumer spending). Additionally, the related outputs were included categories of employment costs and taxes from enterprises.
SALE
NET EXPORT
Net exports are one of the main components of GDP (apart from consumption, government spending and investment). A characteristic feature of the pharmaceutical sector is its high import intensity and generating a trade deficit.
The Strategy for Responsible Development (SOR) indicates foreign expansion as one of the main strategic areas. This area can be taken into account by measuring revenues from export sales.
At the same time, SOR emphasizes the need to support enterprises that create high added value. In order to take into account the added value created in the economy, it is necessary to include the value of imported purchases as reducing the value of domestic production. It is also possible to expand the indicator to separately track the import of raw materials, materials, semi-finished products and the import of goods, to more precisely track the added value.
This also allows for a better consideration of the sector's impact on GDP, which is composed of net exports (the value of exports minus the value of imports for the entire economy).
DATA
Pharmaceutical sector companies report a total of PLN 244 million in income tax annually. It is worth noting that companies with Polish capital have significantly increased their contribution to the tax base over the years under review.
We consider the contribution of pharmaceutical sector companies to the tax base based on the CIT tax. The PIT tax paid by companies on behalf of employees also contributes to the tax base, but due to its collinearity with the amount of remuneration costs, its inclusion would result in double rewarding of companies for the number of employees.
Measuring the amount of corporate tax reported allows for taking into account the specificity of companies from special economic zones. By establishing appropriate weights for CIT-related indicators and investment-related indicators, it will be possible to take into account the efficiency of applied tax incentive policies.
SALARIES
The best measure of the direct impact of a company's activity on the economy through the employment channel is the level of costs related to salaries. Unfortunately, the Accounting Act (Act) allows for the preparation of a profit and loss account in two variants: calculation and comparative. Only the comparative variant separates the costs of salaries and social insurance and other benefits from the operating costs.
The pharmaceutical sector employs, on average, more highly qualified and better paid workers. Additionally, it supports the creation and development of human and social capital, which are areas that influence the achievement of SOR goals.
In order to more strongly promote investment in human capital, it is possible to isolate an indicator limited to employees with higher education or at least a doctoral degree.
Taking into account social security contributions additionally rewards companies that do not use so-called "junk contracts" in their operations (additional variable to be selected in PharmaRank).
Selection of weights
The methodology for assigning weights was developed based on the DELab UW study (2017). The weight values come from normalizing the multipliers of individual influence channels. Normalized value ranges were determined for the influence channels. The category weight is the share in the sum of the means of the ranges of individual multipliers multiplied by 100. The analysis of influence channels was based on:
- NBP NECMOD model: macroeconometric model, multi-equation model reflecting the structure of the Polish economy and its links with the world economy,
- and NBP macroeconomic research (see sources of recommendations).
Co dalej?
The ranking provides the possibility of flexible inclusion of additional categories. Including the value of chemical and biochemical production, including direct sales values of drugs that meet the rules of origin of a Polish product (data currently unavailable) and excluding from the ranking companies in the case of which the value of import to sales exceeds 35% would allow for a more accurate estimate of the actual importance of a given pharmaceutical company for the Polish economy.
If the government administration is interested in the PharmaRank application, it will be possible to further develop it and even more accurately reflect the real impact of individual companies on the Polish economy.
However, this will require all companies in the pharmaceutical sector to be subject to the reporting obligation. Indicators enabling a more detailed analysis of the macroeconomic impact of individual pharmaceutical companies are described in detail in the DELab UW presentation (2017).
Basing the analysis on more detailed data will allow the use of more advanced parameterization methods. Machine Learning techniques (clustering algorithms) will be used to search for similar companies and assign them to individual segments.
RECOMMENDATION SOURCES
DELab UW (2015). Macroeconomic aspects of the importance of the pharmaceutical sector for the Polish economy.
DELab UW (2017). The concept of measuring the importance of pharmaceutical companies in the Polish economy.
National Bank of Poland (2009). The new macroeconometric model of the Polish economy. Working Paper, 62.
NBP (2010a). Re-estimation of the quarterly NECMOD model of the Polish economy.
NBP (2010b). State expenditure multiplier and estimates of its size for Poland. "Materials and Studies", 246.
NBP (2012a). Re-estimation of the quarterly model of the Polish economy NECMOD.
NBP (2012b). Fiscal multipliers and factors of growth in Poland and the Czech Republic in 2009. Working Paper, 117.
National Bank of Poland (2015). Rock multipliers in the credit-constrained model. "Materials and Studies", 318.