Introduction
In recent years, a growing number of mobile applications have emerged offering services under the “Buy Now, Pay Later” (BNPL) label. Behind these applications are companies operating within an economic sector that was formally regulated by law only about six years ago. While this sector is relatively new, it engages in a long-standing practice, now powered by new tools that have enabled it to expand rapidly within a short period of time.
Consumer lending, the provision of loans to individuals to enable them to purchase goods and services, is not new. Nor is the effort to use data to assess individuals’ ability to meet their financial obligations. What is now referred to as “alternative data” reflects a shift away from relying solely on traditional banking data, which banks have historically used to determine creditworthiness and lending limits for individuals and businesses.
Digital technologies have transformed these practices by making consumer lending more accessible and easier to obtain. They also enable the collection and analysis of large volumes of detailed personal data. Advanced analytical tools, including those based on artificial intelligence, can process this data at scale, leading to the automation of decisions about individuals’ eligibility for consumer lending.
This paper argues that the use of digital technologies to collect and process alternative data for consumer lending goes beyond merely facilitating access or expanding services. In the Egyptian context in particular, this use carries significant and potentially harmful implications. These impacts extend from individuals’ daily lives and consumption patterns to broader questions of social justice and the structure of the economy as a whole. The paper captures this central argument through the concept of “investment in poverty.” This concept highlights not only the exploitation of rising poverty levels to generate profit, but also the active expansion and deepening of poverty in ways that serve the interests of the capitalist market and sustain its growth.
The paper begins by defining alternative data and its intended purposes, before examining its sources and the tools used to process it. It then considers the potential for bias in both the collection and analysis of such data. In addition, the paper provides an overview of consumer lending in Egypt and the role of alternative data within it, including the sector’s rapid growth and the scope and limitations of its current legal framework.
The paper also critically examines how the concepts of financial inclusion and social justice are used in promoting consumer lending in Egypt. It further explores structural factors specific to the Egyptian context that intensify the negative consequences of using alternative data and artificial intelligence in this sector. These include the nature of the labor market and income sources, fluctuating consumption patterns, and the fragility of the economic structure. Bringing these elements together, the paper presents consumer lending as a practical manifestation of “investment in poverty” in its various dimensions.
What Is Alternative Data and What Is Its Purpose?
The term alternative data can refer to different types of data depending on the context in which it is collected and processed. In general, it describes data that serves as a substitute for information traditionally collected within a given context. Its primary purpose is to enable activities that rely on data which, under conventional conditions, may be unavailable, irregular, or insufficiently representative to support reliable conclusions.
For example, traditional consumer lending typically relies on banking records, including an individual’s deposit and withdrawal history. Such methods cannot be applied to individuals who do not have bank accounts or who do not regularly use formal banking services for their day-to-day financial transactions. This situation reflects the reality for a large proportion of people in Egypt.
Alternative data is also intended to expand beyond the types of information traditionally considered. This broader data collection aims to provide a more detailed and comprehensive picture of individual behavior, one that is often presented as more reliable than conventional data sources.
In financial contexts, for instance, alternative data may reveal indicators of an individual’s ability and willingness to meet recurring financial obligations—indicators that may not be visible through bank records alone. The availability of larger volumes of data also enables the use of analytical tools that depend on data abundance, particularly artificial intelligence (AI) systems.
In the Egyptian context, the size of the market, driven by a large population, contrasts sharply with the limited purchasing power of the majority. This gap has widened in recent years due to the rapid depreciation of the Egyptian pound, further reducing purchasing power. As a result, even social groups that were previously among the highest consumers of goods and services are no longer able to maintain their usual levels of consumption. This has increased reliance on consumer lending as a means of mitigating market contraction.
At the same time, limited purchasing power is accompanied by restricted access to traditional banking services for much of the population. Consequently, conventional lending models that rely on banking transaction data are unable to effectively include large segments of potential consumers. This limitation has created a strong incentive to adopt alternative data as a basis for expanding access to consumer credit.
Sources of Alternative Data and Tools for Its Processing
A report by the Egyptian Banking Institute, affiliated with the Central Bank of Egypt, indicates that alternative data includes “mobile phone usage patterns, digital wallet transactions (including those provided by mobile network operators), rent and utility payments, online shopping behavior, and even indicators of engagement on social media platforms.”
Taken together, this data is used to construct a profile of individuals through which their level of responsibility, the soundness of their financial decisions, and ultimately their ability and willingness to meet the financial obligations associated with a given loan can be assessed.
Accessing this data requires service users to provide some of it directly, either in paper or digital form. Other data is obtained from third parties, as well as from individuals’ social media accounts and the data collected and stored by mobile applications on their devices. In practice, the volume of alternative data used for consumer lending far exceeds that used in traditional credit assessment methods. As a result, processing this data typically relies on AI models capable of handling large-scale data analysis.
Risks of Bias in the Collection and Processing of Alternative Data
AI models tend to reproduce the biases embedded in the datasets on which they are trained. These biases can take multiple forms, including statistical bias, conceptual bias, and biases rooted in prejudice and social stereotypes. In the context of alternative data used for consumer lending, statistical bias often arises from unequal levels of data availability across different social groups.
For example, patterns of mobile phone usage vary significantly across social groups. The same applies to the use of digital wallets, the availability of records for rent and utility payments, reliance on online shopping, and levels of engagement on social media. These variations reflect broader social and economic inequalities.
As a result, there are substantial disparities in how individuals are represented in available datasets. This can lead to the exclusion of certain individuals—not because they are ineligible, but because the data available about them falls below the minimum threshold required for algorithmic assessment to produce “acceptable” results regarding their eligibility for loans of a given size.
In addition, the collection and processing of alternative data involve conceptual biases. Concepts such as stable employment, steady income, and “responsible” financial behavior are defined through specific indicators that may reflect the realities of certain social groups while failing to capture the lived conditions of others.
This problem is compounded by the fact that available data is often partial or incomplete, and therefore may not accurately reflect reality. As a result, conventional indicators may fail to capture the nature of work, income sources, and financial practices of many individuals. This can expose them to discrimination and exclusion from consumer lending opportunities, even when they are, in practice, capable of meeting their financial obligations.
Moreover, AI models are particularly vulnerable to biases rooted in prejudice and prevailing social stereotypes. A significant portion of the data they rely on consists of past evaluation outcomes produced by human decision-makers. These historical assessments often reflect, whether consciously or unconsciously, existing biases and stereotypes. Over time, this results in statistical patterns where certain groups are disproportionately associated with negative outcomes, while others are more likely to receive positive evaluations.
For example, historical data may reflect disproportionately negative assessments of women, based on entrenched stereotypes that question their financial responsibility or consumption decisions. Similarly, negative evaluations may be associated with individuals’ marital status, such as being unmarried or divorced, number of children, housing status, or other personal characteristics. When such biased data is used to train AI systems, these systems are likely to replicate and reinforce these patterns of discrimination.
The Growth of Digitally-Driven Consumer Lending in Egypt
“Buy Now, Pay Later” has become the primary marketing slogan for new consumer lending services in Egypt. These services rely heavily on mobile applications to reach as many individuals as possible within their target groups. This shift toward mobile-based delivery is not only a strategic choice by companies operating in the consumer lending sector, but also a direction actively supported by public authorities.
On 13 October 2024, the Financial Regulatory Authority (FRA) issued a decision to suspend the acceptance of new applications for establishing consumer finance companies through traditional means for a period of one year. The decision explicitly exempted entities seeking to operate in this sector using financial technology (fintech). Although the decision stated that it was not subject to renewal, the FRA extended its implementation for an additional year in October 2025.
The legislative authority enacted Law No. 18 of 2020—amended in 2022—to regulate the activities of consumer finance providers. The law places these providers under the supervision of the Financial Regulatory Authority, which is responsible for enforcing the law and issuing related administrative decisions. It also mandates the establishment of a representative body for the sector: the Egyptian Federation of Consumer Finance.
According to recent televised statements by the head of the Federation, the consumer finance market is currently growing at an annual rate of 50 percent. The total volume of financing provided by companies in this sector increased within a single year (2024–2025) from 47 billion Egyptian pounds to 74 billion pounds by the end of October 2025. During the same period, the number of clients rose from 5 million to 9 million, while default rates on financial obligations ranged between 2 percent and 4 percent.
These figures differ from those reported in media sources based on official data from the Financial Regulatory Authority. However, the latter also reflects the rapid expansion of the consumer finance market in recent years, as illustrated below:
| Year | Total Financing (EGP billions) | Growth Rate |
| 2020 | 8.4 | |
| 2021 | 17.2 | 104.8% |
| 2022 | 29.8 | 67.4% |
| 2023 | 47.3 | 58.7% |
| 2024 | 61.3 | 29.6% |
What the Law Regulates and What It Overlooks
The provisions of the law regulating the establishment and operation of consumer finance providers set out the requirements that entities must meet in order to obtain a license to operate in this sector. These include, among other conditions, the requirement to have “the necessary equipment, technological infrastructure, and information systems to carry out the activity in accordance with the requirements set by the Authority.” The law also addresses in detail the financial solvency requirements for companies and includes safeguards to prevent their operations from being used for money laundering or terrorist financing.
However, only one article of the law addresses clients’ rights, Article 5, which states:
“Consumer finance companies, their managers, advisors, consumer finance providers, and their employees shall maintain the strict confidentiality of their clients and shall not disclose any information about them or their transactions to third parties without their prior written consent and within the limits of such consent, except in cases where specific information must be disclosed in accordance with applicable laws.”
The law does not establish any explicit obligations for regulated entities regarding the protection of personal data in the context of collecting, storing, processing, or sharing such data. It is important to note that references to “strict confidentiality” and non-disclosure are broad and vague, and do not amount to meaningful data protection safeguards.
In particular, these provisions do not include any guarantees against the misuse of personal data. The law also does not clarify whether the entities it regulates are subject to Egypt’s Personal Data Protection Law, even after its amendment in 2022. Under that law, entities such as consumer finance companies should fall within its scope, as they are not covered by the exemption granted to institutions supervised by the Central Bank.
Moreover, the law does not address the use of software or artificial intelligence (AI) models in the collection and processing of customer data. It also fails to provide any safeguards against bias or discrimination in decision-making processes, whether resulting from digital technologies or human involvement.
Consumer Lending, Financial Inclusion, and Social Justice
The Financial Regulatory Authority provides an informational brief on consumer finance on its official website. This brief describes consumer finance as “one of the main tools for achieving social justice, as it enables middle- and low-income groups to access financial services instead of restricting them to large companies and financially privileged individuals.” It also presents consumer finance as a component of “financial inclusion,” which it describes as “one of the pillars of the Sustainable Development Goals adopted by the United Nations General Assembly at its sixty-ninth session.”
This narrative stands in tension with the core meaning of social justice, which is fundamentally concerned with the fair distribution of wealth and the reduction of income inequality. It overlooks the reality that low incomes—where individuals are unable or barely able to access essential goods and services necessary to sustain a basic standard of living—are themselves a manifestation of social injustice.
Providing an alternative that imposes additional financial burdens in order to access these essential goods and services constitutes a further form of inequality. Individuals with limited or low incomes effectively pay higher prices for the same goods and services due to the costs associated with borrowing.
In addition, users of “Buy Now, Pay Later” services bear a hidden cost linked to the use of alternative data in these systems. This cost arises from being compelled to provide sensitive personal data to entities that are not meaningfully bound to protect it under the current legal framework. Furthermore, their financial decisions become increasingly shaped by digital systems that process their data and assess their behavior without any safeguards to prevent bias or discrimination. This creates a risk of unfair treatment based on identity characteristics or behavioral and psychological profiling.
On the other hand, financial inclusion is not inherently linked to any meaningful concept of social justice. Rather, it is primarily concerned with expanding markets to absorb the limited incomes available to the majority, many of whom have been impoverished by the very dynamics of these markets. This expansion serves to protect capital investments that would otherwise be threatened by market contraction resulting from the inability of low- and middle-income groups to afford the goods and services offered.
Challenges of Using Alternative Data in the Egyptian Context
Labor Market Dynamics and Sources of Income
The Egyptian economy can currently be classified among what are often referred to as peripheral economies. These are economies whose degree of integration into the global system fluctuates depending on the shifting needs of that system for expansion or contraction. The role of such economies is to absorb and mitigate the effects of these fluctuations on more central economies.
As a result, the Egyptian economy is characterized by continuous volatility, oscillating between periods of growth and contraction due to external pressures, compounded by internal factors that intensify these effects. Overall, this environment does not allow for the development of stable and consistent economic practices.
This instability makes it difficult to sustain stable employment relationships and increases the cost of providing secure jobs. In recent years, the Egyptian economy has also experienced multiple shocks, including significant currency devaluation and rising inflation, which have sharply increased the cost of living. These conditions place workers under pressure to accept any available employment opportunities, regardless of whether basic standards of job security are met.
Consequently, labor market practices in Egypt are largely characterized by informality, that is, a lack of compliance, in one form or another, with legal frameworks governing employment relationships. Since work is the primary, and often the only, source of income for the vast majority of Egyptians, employment relationships should, in principle, be the main source of data relevant to assessing individuals’ financial capacity. However, due to the widespread informality of these relationships, such data is either not recorded at all or, when recorded, does not accurately reflect reality.
For example, many individuals work in what are effectively permanent positions, while their official employment status is recorded as temporary contractual work. Conversely, official records may list basic monthly salaries that differ from actual earnings, often to reduce social insurance obligations.
In addition, a significant proportion of the population works within the so-called informal economy, which exists entirely outside formal record-keeping systems—whether paper-based or digital. This includes domestic workers, street vendors, artisans, and many others. For these groups, reliable data on income and on most financial transactions is largely unavailable.
Fluctuating Consumption Patterns
One of the key types of alternative data used in consumer finance, particularly “Buy Now, Pay Later” services, is the consumption behavior of targeted groups. This data is typically derived from purchase histories recorded through mobile applications and online platforms. However, for many individuals, such records do not provide a complete or accurate picture of their actual consumption patterns.
At the same time, there is another critical dimension that digital assessment tools struggle to capture accurately: the inherently fluctuating nature of consumption patterns in Egypt. These fluctuations result from a combination of factors, some economic, but many rooted in social and cultural dynamics.
On a recurring annual basis, consumption levels in Egypt tend to rise during religious seasons, such as major holidays. In addition, certain significant social events, most notably marriage preparations, can lead to sharp increases in spending on goods and services. These expenses are often difficult to cover without resorting to borrowing, and in many cases, this creates a high risk of default.
In such situations, social and cultural expectations can outweigh considerations of financial responsibility. This dynamic has contributed to the emergence of a broader social phenomenon known as al-gharimat, women who have taken on debt to meet household or family-related expenses beyond their capacity to repay, often resulting in legal prosecution and imprisonment.
The use of consumption data, borrowing histories, and even records of legal action or imprisonment due to non-payment in assessing individuals’ eligibility for consumer loans can lead to the exclusion of many people. Their recorded consumption patterns may be interpreted as evidence of financial irresponsibility. While this assessment may hold in extreme cases, it fails to account for seasonal and socially driven consumption behaviors. AI systems, in particular, are not equipped to distinguish between these contexts.
It is also important to note that seasonal fluctuations in consumption tend to appear more pronounced among lower-income groups, while they may be less visible among higher-income individuals whose spending levels are generally higher and remain within their financial capacity. Ignoring the underlying causes of these fluctuations, therefore, constitutes a form of structural bias and discrimination against low- and middle-income groups.
The Fragility of the Economic Structure
The fragility of an economic structure refers to its limited ability to absorb sudden shocks arising from unexpected internal or external conditions. This fragility leaves individuals without meaningful protection in times of crisis, exposing them to rapid and severe disruptions in income, spending capacity, and their ability to maintain basic living standards.
Within such a fragile economic context, taking on additional long-term financial obligations represents a significant risk for low- and middle-income groups. In recent years, a combination of internal and external factors has led to a sharp depreciation of the Egyptian pound, losing roughly half of its value. This decline has had devastating effects on these groups. In the absence of sufficient safeguards to protect spending on essential needs, the resulting loss in purchasing power has forced many individuals to forgo basic necessities.
Amid the expansion of “Buy Now, Pay Later” services, many people in Egypt have turned to borrowing in an attempt to restore living standards eroded by high inflation in recent years. Loan repayments are typically drawn from the small margin of income that exceeds minimum subsistence needs. This means that any new inflationary shock would likely push a large number of borrowers into default.
The consequences of such widespread default could be devastating for many individuals and households. Moreover, the growing reliance on alternative data and artificial intelligence (AI) systems increases the likelihood of this scenario, as these systems are inherently unable to account for the structural fragility of the economy and its impact on individuals’ capacity to meet financial obligations.
At a systemic level, the risk of mass default in a fragile economic environment raises the possibility that the entire consumer lending sector could become a “ticking time bomb,” the collapse of which could exacerbate any future currency crisis. The scale of the sector’s recent growth and its projected expansion in the coming years means that a significant conversion of outstanding obligations into non-performing debt would have serious repercussions for the Egyptian economy.
This risk is further heightened by the increasing tendency of companies in the sector to securitize their loan portfolios. In simple terms, this means that financial and investment institutions, such as banks, investment funds, and pension funds, become exposed to the risks associated with consumer debt. As a result, they too face potentially significant losses in the event of widespread default.
Investment in Poverty
There is perhaps no clearer or more direct definition of poverty than the inability of an individual or household’s income to cover the cost of the goods and services required for their standard of living. Unlike financing activities that support the establishment or operation of profit-generating enterprises, which can be understood as investments in individuals’ aspirations to increase their income, consumer finance is better understood as an investment in people’s need to obtain essential goods and services that their incomes cannot adequately cover.
The profits generated by consumer finance providers are derived from the additional costs that users are compelled to pay. These costs include interest rates designed to hedge against expected inflation over the repayment period, as well as service fees and profit margins, all of which are added on top of the original price of the goods and services—prices that may themselves already exceed what would be expected under normal market conditions.
In its simplest terms, this activity can therefore be described, to a large extent, as an investment in poverty. The rapid expansion of this sector in Egypt provides a clear illustration of this dynamic, as it has been driven by a sharp and rapid depreciation of the national currency, which effectively halved the purchasing power of large segments of the population.
Some may argue that consumer finance represents an investment in individuals’ consumption aspirations, allowing them to access goods and services beyond their immediate financial means but within their expected future capacity to pay. While this argument may hold in other contexts, it applies only to a limited segment of beneficiaries in the Egyptian case.
As an indication, the following figures show the distribution of consumer lending across major sectors:
| Sector | Share of Consumer Finance |
| Cars | 20.5% |
| Electronics | 17.1% |
| Electrical and household appliances | 15.9% |
| Consumer goods via financing cards | 12.4% |
| Home finishing | 4.1% |
| Clothing, footwear, bags, watches, jewelry, and eyewear | 4% |
| Furniture and home furnishings | 2.7% |
At first glance, these figures may appear to contradict the argument above. However, it is important to note that they reflect the distribution of financing value, not the number of individuals benefiting from each category. When adjusted for the cost of individual goods, the data suggest that the vast majority of beneficiaries use consumer finance to purchase everyday goods with financing cards. By contrast, those using such financing to purchase cars represent only a very small minority, likely numbering in the thousands among millions of clients. The second largest categories, household appliances and electronics (such as mobile phones and computers), are increasingly essential to maintaining a basic standard of living, particularly for middle- and lower-income groups.
There is another dimension to “investment in poverty”: the expansion of consumer lending at scale contributes directly to the deepening of poverty. On one hand, it imposes additional financial burdens on consumers beyond the original price of goods and services, including service costs, provider profit margins, and inflation-adjusted interest. Where loan portfolios are securitized, additional profit margins are introduced, some absorbed by providers, but others passed on to consumers as further costs.
Moreover, when consumer lending reaches a scale large enough to influence the broader market, its effects extend beyond direct users. It generates artificial demand for goods and services, contributing to price increases and, consequently, higher inflation.
Given the size of the consumer finance market, it may also lead to a shift of savings deposited in banks toward consumption. While banks can recover these deposits through securitization, transforming savings that once generated returns for depositors into debt instruments that generate interest income, depositors themselves may effectively transition from creditors to debtors.
In practice, this reflects one of the less visible dimensions of financial inclusion. In this form, financial inclusion functions as a mechanism for absorbing as much liquidity as possible, from small amounts of cash circulating in daily transactions or from relatively costly small deposits, into large-scale, aggregated financial systems within the banking sector, making these resources available for capital investment markets.
Conclusion
Within the framework outlined in this paper, alternative data and digital technologies play a central role in safeguarding investments in consumer finance and reducing associated risks. In doing so, they become integral to a broader system designed to extract the limited financial resources that remain in the hands of lower-income groups and channel them into the market to generate further profit.
In the Egyptian context, the impact of this system is intensified by the fragility of the economic structure, the informality of the labor market, and the weakness of legal and regulatory frameworks governing data protection and the prevention of discrimination. These conditions expose individuals to increasing financial burdens and to digital practices that lack transparency and accountability.
Addressing this phenomenon, therefore, requires a comprehensive rethinking of data protection policies, stronger safeguards for algorithmic fairness, and clear alignment between financial inclusion initiatives and principles of social justice. Such measures are essential to ensure that digital technologies do not become tools for reproducing and deepening poverty, but instead contribute to addressing it.