Overcoming data challenges in the financial services sector

Importance of the financial services sector

The financial services industry plays a significant role in global economic growth and development. The sector contributes to the creation of a more efficient flow management of savings and investments and enhances risk management of financial transaction activities for products and services. Institutions such as commercial and investment banks, insurance companies, non-banking financial companies, credit and loan companies, brokerage firms, and trust companies offer a wide range of financial services and distribute them in the marketplace. Some of the most common financial services are credits, loans, insurances, and leases, distributed directly by insurance companies and banks, or indirectly via agents and brokers.

Limitations and challenges in data availability

Due to the important role of financial services in the global economy, it is expected that the financial services market is professional and highly developed, also in terms of data availability. Specifically, a well-designed database is expected to be available, where a wide range of information is presented and can be collected regarding certain industries. However, reality does not meet these expectations.

Through assessments of various financial service markets, it has been observed that data collection is a challenging process. Several causes contribute to this situation. Lack of data availability or poor data availability, data opacity, consolidated information from market or annual reports, as well as different categorization schemes of financial services are some of the most significant barriers. Differences in the legal framework among countries have a major impact on the entry and categorization of data. A representative example that applies in this case is the different classification schemes and categorization of financial services across countries. Specifically, EU countries are obligated to publish data of financial service lines under a certain classification scheme and pre-defined classes, which in many cases differ from the classification schemes or classes of non-EU countries, contributing to an unclear, inaccurate overview of the market. The identification and understanding of each classification scheme are necessary to avoid double counting and overlapped data. In addition, public institutions often publish data, revealing part of the market and not presenting the actual market sizes. Lastly, it has also been observed that some financial services have different definitions across countries, which influences the complexity of the data collection and assessment of the financial services market.

Need for a predictive model

In order to overcome the challenges of data inconsistency and poor, limited, or non-existent data availability and to create an accurate estimation of the financial services market, it is necessary to develop a predictive model that analyzes a wide range of indicators. A characteristic example is the estimation of the global financial services market conducted by The World Bank. An analysis model, based on both derived and measured data information, was created to address limited data input challenges.

An analysis model for the assessment of the financial services markets, created by Hammer, takes into consideration both the collection of qualitative and quantitative data from several sources as well as predictive indicators. In previous assessments of certain financial services markets, data information was collected from publications, articles, and reports from public financial services research institutions, country financial services associations, association groups, and private financial services companies. Field experts' opinions also constituted a significant source of information. The model included regression and principal component analysis, where derived data were produced based on certain macroeconomic factors (such as country population, GDP, GDP per sector, unemployment rate), trade indicators, and economic and political factors.

The selection of the indicators and analysis model depends on the type of financial service product and the relative market that we want to assess. In addition, based on model analysis, it is possible to identify and validate correlations between a set of predictive indicators that have been considered potential key drivers of the specific markets.

To conclude, it is possible to identify the sizes of the financial services markets, with the support of an advanced predictive analysis model that can enable and enhance comparability and consistency of data across different markets and countries.