Unit 9 Lenders and borrowers and differences in wealth

9.9 Borrowers and lenders: A principal–agent problem

Before you start

To understand this section, you will need to know about principal–agent problems and incomplete contracts, which were introduced in Unit 6. If you are not familiar with these concepts, you should read Section 6.6 before proceeding further.

principal–agent relationship, principal–agent problem
A principal–agent relationship or problem exists when one party (the principal) would like another party (the agent) to act in some way, or have some attribute, that is in the interest of the principal, and that cannot be enforced or guaranteed in a binding contract. See also: incomplete contract.
credit market constrained
A description of individuals who are limited in how much they can borrow or can borrow only on unfavourable terms. See also: credit market excluded.
credit market excluded
A description of individuals who are unable to borrow on any terms. See also: credit market constrained.

In this section, we will explain how the relationship between the lender and the borrower is a principal–agent problem similar in many ways to the relationship between the employer and employee studied in Unit 6. The lender is the ‘principal’ and the borrower is the ‘agent’. The principal–agent model explains why many people who want to borrow are not able to do so, or can only borrow limited amounts or at high interest rates.

Credit market exclusion when loans may not be repaid

In the previous sections, we simplified the analysis of borrowing and lending in three ways. First, any loans taken on would be repaid. Second, there was a given interest rate; we did not explain where that came from. And third, borrowing funds was possible even for a person with no initial wealth.

None of the three assumptions fits actual credit markets where lenders decide on the interest rate to be charged, loans may not be repaid, and, for that reason, some who would like to borrow are not as fortunate as Julia. Therefore, many prospective borrowers are either credit market constrained, that is, they:

  • are limited in how much they can borrow, or
  • can borrow only at very high rates of interest (like those taking out payday loans in New York at the beginning of the unit).

Alternatively, they are credit market excluded, that is, they are simply unable to borrow at all.

Figure 9.15 gives an idea of the limitations that people face in credit markets in the US (the data is for 2019, so as to avoid the unusual difficulties faced by people during the COVID-19 pandemic). To understand what the figures mean, think about a person whose job is terminated. In 2019, somewhat more than one in five people would be unemployed for more than six months. The table shows that almost one-third of people could not ‘cover expenses by any means’ including borrowing, for even just three months. More than half were borrowing at the very high rates of interest charged on credit cards. Not surprisingly, those who were credit market constrained or excluded tended to be poor. But even those with substantial incomes (over $100,000) sometimes found that they could not borrow as much as they would have liked.

This bar chart shows different types of market exclusion and credit constraints in the US. About 50% of people are borrowing at credit card interest rates between 16% and 20%. About 30% of people cannot cover three month’s expenses by any means if primary income source is lost. Over 15% of people with an income greater than $100,000 are credit market excluded or constrained. About 30% of people with an income between $40,000 and $100,000 are credit market excluded or constrained. About 50% of people with an income less than $40,000 are credit market excluded or constrained.

Figure 9.15 Indicators of credit market exclusion and credit constraints in the US in 2019.

Board of Governors of the Federal Reserve System. 2020. Report on the Economic Well-Being of U.S. Households in 2019.

The ‘How economists learn from facts’ box presents additional evidence on the credit market limitations that people face.

How Economists learn from facts Who is credit constrained?

One way to estimate the prevalence of credit constraints is through surveys. This is how the US Federal Reserve Board got the data shown in Figure 9.15.

Although useful, survey data has limitations: people may not remember their credit market experiences accurately and may not be able to accurately answer hypothetical questions about whether they would succeed in getting a loan if they attempted to borrow.

For an explanation of treatments and controls in experiments, read ‘How economists learn from facts: Laboratory experiments’.

Another way to investigate credit constraints is to run experiments, in which one group is randomly assigned to receive an intervention (often called a ‘treatment’), with the other group not receiving the intervention and referred to as the control group. Because the assignment of an individual to one of the groups is random and because other influencing factors are held constant, average differences in behaviour between the groups are interpreted as caused by the treatment.

If a person is credit constrained, being given a sum of money might change their behaviour. For example, they might start up a company. On the other hand, if they were not credit constrained and had thought it would be profitable, they would have already taken out a loan. Credit constraints can therefore be identified by running experiments in which some individuals are given sums of money (the treatment group), and others are not.

In one experiment, the Nigerian government selected the best business plans from those that were submitted to them. Within this pool of the best plans, they randomly assigned some to receive a cash grant worth, on average, around 8 million naira (which was approximately equivalent to US$50,000 when the study was published). The winning applicants expanded their existing businesses, by hiring more workers and purchasing more capital goods.1 They could have used the grant to lend at the going interest rate, but instead they chose to expand their business. Before the grant, they were unable to undertake the business expansion. This implies that they were previously constrained in their ability to borrow. A similar experiment in Sri Lanka randomly assigned small businesses a grant worth 10,000 LKR (which was approximately equivalent to US$100 when the study was published). The firms that received the grants subsequently became much more profitable,2 implying that credit constraints were limiting their ability to make profitable investments.

In Unit 1, we described the natural experiment that occurred because, at the end of the Second World War, Germany came to be subject to two very different political and economic systems: Communist Party rule and central economic planning in the East, and democracy and capitalism in the West. We then considered the differences in their institutions as the cause of the different paths of economic growth of the two parts of Germany.

Experiments are often considered as a ‘gold standard’ for identifying causal effects. However, experiments are often difficult to implement because they can be expensive, difficult to implement logistically, and sometimes raise ethical questions. Because of this, economists often use ‘natural experiments’, where a naturally occurring event is considered ‘as good as randomly assigned’.

In the case of credit constraints, the arrival of an unexpected inheritance can be considered as good as random.

In Britain, it was estimated that an inheritance of approximately £5,000 (which was approximately equivalent to US$9,000 when the study was published) doubled young people’s likelihood of starting up a new business,3 implying that they were previously constrained in their ability to borrow money to do so. It has also been found that after receiving an inheritance, individuals are more likely to transition into self-employment and to increase the scale of their operations if already self-employed.4

Another natural experiment that can be used to identify whether individuals are credit constrained is automatic increasing of borrowing limits on a credit card, after the card has been held for some time. Evidence from the US5 suggests that many individuals are indeed credit constrained, since an increase in their credit limits, which allows them to borrow more, results in them choosing to increase their borrowing.

Exercise 9.9 Estimating credit constraints

List some of the advantages and limitations of using each of the following methods to learn about who is credit restrained. (You may find the websites and papers linked to in the ‘Who is credit constrained?’ box helpful.)

  1. surveys of individuals
  2. field experiments (for example, randomly selecting some people to receive a lump sum of money)
  3. natural experiments (for example, inheritances and increasing borrowing limits on a credit card).

Lending is risky

Borrowing and lending is a principal–agent relationship because the lender (the principal) faces the risk of not being repaid, and the extent of that risk is determined by the borrower (the agent), not the lender.

A loan is made now and has to be repaid in the future. Between now and then, unanticipated events beyond the control of the borrower can occur. If the crops in Chambar, Pakistan were destroyed by bad weather or disease, the moneylenders would not be repaid. If the skills you have invested in using your student loan become obsolete, the loan may not be repaid. Unanticipated events like these are known as unavoidable risks. The interest rate set by a bank or a moneylender would be greater if the default risk due to unavoidable events was greater.

But lenders face two further problems. When loans are taken out for investment projects, the lender cannot be sure that a borrower will exert enough effort to make the project succeed. Moreover, often the borrower has more information than the lender about the quality of the project and its likelihood of success. Both of these problems arise from the difference between the information the borrower and the lender have about the borrower’s project and actions.

Think about a borrower who secures a loan to start a new business. The project doesn’t succeed because the borrower made too little effort or because it just wasn’t a good project. If the borrower has not put any of their own money into the project, it is the lender, not the borrower, who loses money if the project fails and the loan is not repaid. If the borrower were using only their own money, it is likely that they would have been more conscientious or maybe not engaged in the project at all.

The borrower may also use the borrowed funds for a much riskier project than the one that they told the lender they would use it for. To illustrate this (with an extreme example), they could simply buy lottery tickets with the money they have borrowed—if one of them pays off, they are rich; if not, the lender does not get repaid.

Just like the employee in Unit 6, who did not always act in the best interest of the employer, the borrower does not necessarily act in the best interest of the lender.

The principal–agent problem between borrower and lender is also similar to the ‘other people’s money’ problem which you can read about in Section 6.3. In that case, the manager of a firm (the agent) makes decisions about the use of the funds supplied by the firm’s owners (the principals), but the owners cannot contractually require the manager to act in a way that maximizes the owners’ wealth, rather than pursuing the manager’s own objectives.

Incomplete contracts

In the case of borrowing and lending, it is often not possible for the lender (the principal) to write a contract that ensures a loan will be repaid by the borrower (the agent). The reason is that if the project fails, or if the borrower is for other reasons without funds when repayment is due, it may be impossible for the lender to secure repayment according to the terms of the loan.

Lenders will attempt to secure repayment of a loan through legal measures, but this will often be difficult if the borrower is poor or declares bankruptcy because the value of their debt exceeds the value of their assets. In the introduction to this unit, we reported an example of a method of improving compliance in car loan repayments—companies install devices that disable the ignition of the car if the repayments are not made as required.

If legal methods fail, lenders may use illegal ones, such as threatening physical violence.

The role of collateral in lending

An individual holds equity in a project or business if some of their own wealth (rather than borrowed funds) is invested in it. There is a second entirely different use of the term, meaning fairness, as in ‘an equitable division of the pie’.
An asset that a borrower pledges to a lender as a security for a loan. If the borrower is not able to make the loan payments as promised, the lender becomes the owner of the asset.

One response of the lender to this conflict of interest in the credit market is to require the borrower to put some of their wealth into the project (this is called equity). The more of the borrower’s own money that is invested in the project, the more closely aligned their interests are with those of the lender. Another common response is to require the borrower to set aside property that will be transferred to the lender if the loan is not repaid (this is called collateral).

Collateral is used in loans for houses (called mortgages) and for cars. For many people (as in Figure 9.15 for the US), these are the only large loans they can get, and that is because the collateral—the house or the car—reverts to the lender if repayments are not made. Securing a loan to cover your consumption needs during a period of unemployment is much more difficult because the unemployed worker is unlikely to have any collateral to offer.

The pawnbroker is a common example of collateral in small-scale lending and borrowing that has existed for thousands of years. The pawnbroker, found today on shopping streets under the slogan ‘cash converter’ or similar, extends a loan to the borrower with a date and amount of repayment specified. And the borrower turns over some item of their property to the pawnbroker, which will be returned to the borrower when the loan is repaid. Items commonly lodged with a pawnbroker—because they can easily be sold—include jewellery, laptops and other electronic equipment, cameras, or valuable household items.

A loan with collateral is called a secured loan because, as long as the collateral (the house or the pawned item) can readily be sold for more than the amount of money owed, the lender is secure. With a secured loan, the lender does not run any substantial risk.

Equity or collateral reduces the conflict of interest between the borrower and the lender. The reason is that, when the borrower has some of their money (either equity or collateral) at stake:

  • They have a greater interest in working hard so that the loan will be repaid: They will try harder to make prudent business decisions to ensure the project’s success.
  • It is a signal to the lender: It signals that the borrower thinks that the project is of sufficient quality to succeed.

The relationship between wealth, project quality, and credit is summarized in Figure 9.16.

This diagram describes the process of obtaining a loan. First, the lender assesses the characteristics of the borrower and the project, based on the wealth of the borrower and the quality of the project (known better to the borrower than to the lender). Second, the lender finds out what the borrower is willing to contribute, which would be the borrower’s equity stake in the project or collateral. Third, the lender uses this information to assess the borrower’s incentives to work hard and to select a high-quality project. Finally, the lender makes a decision on whether or not to provide the loan.

Figure 9.16 Wealth, project quality, and credit.

Two principal–agent relationships: Labour markets and credit markets

The similarity of the credit market and the labour market as principal–agent relationships is illustrated in Figure 9.17.

Actors Conflict of interest over Enforceable contract covers Left out of contract (or unenforceable) Result
Labour market (Unit 6) Employer
Wages, work (quality and amount) Wages, time, conditions Work (quality and amount), duration of employment Effort under-provided; unemployment
Credit market (Unit 9) Lender
Interest rate, conduct of project (effort, prudence) Interest rate Effort, prudence, repayment Too much risk, credit constraints

Figure 9.17 Principal–agent problems: the credit market and the labour market.

Figure 9.18 connects the labour and credit markets. It illustrates how the credit and labour markets influence the relationships among the groups of lenders and borrowers, and employers and employees.

Starting at the upper left of the figure, wealthy individuals can use their wealth to purchase capital goods to become employers, and they can also lend to others. Among the less wealthy, there will be some successful borrowers who can, as a result, also become employers. Those with little wealth cannot borrow much: they are credit market excluded or can borrow only where a house provides the collateral for a mortgage, for example, or access limited amounts at high interest rates. They must seek work as employees. Employers hire employees from among the less wealthy, with some remaining unemployed (due to the workings of the labour market that we study in Unit 6).

Horizontal arrows indicate a principal–agent relationship. Lenders and employers are the principals in the figure; their common red colour indicates this similarity. Agents—successful borrowers, and employees—are coloured green to distinguish them from would-be agents (credit market excluded and unemployed) who are coloured purple. Even if you are an agent lucky enough to be in one of the green boxes, the principal can put you back in the purple box just by refusing to deal with you. This is why lenders and employers have power over borrowers and employees.

This flowchart shows that in an economy there are wealthy individuals and prospective employees. Wealthy individuals can act as lenders and lend to prospective borrowers: some of whom will become actual borrowers, and some of whom will be excluded borrowers. Wealthy individuals and actual borrowers can become employers. Prospective employees who are hired will become employed; those unhired will be unemployed.

Figure 9.18 The credit and labour markets shape the relationships between groups with different endowments.

Figure 9.18 helps us understand why some people end up as principals (employers, for example) while others end up as agents (employees). If one is wealthy, one can be both a lender and an employer.

Question 9.13 Choose the correct answer(s)

Read the following statements about the principal–agent problem and choose the correct option(s).

  • A principal–agent problem exists in the credit market if the lender may not be repaid.
  • The principal–agent problem can be resolved by writing a binding contract for the borrower to exert full effort.
  • One solution for the principal–agent problem in the credit market is for the borrower to provide equity.
  • The principal–agent problem leads to credit market constraints and exclusion.
  • A principal–agent problem exists only if there is something that the borrower may do or be that the lender does not know and cannot control; not being repaid could take place even if there were no such problems of asymmetric information and incomplete contract.
  • The principal–agent problem exists because one cannot write a binding contract for full effort.
  • Equity implies that the agent has more to lose if the project fails, reducing the difference in the incentives between the principal and the agent.
  • This occurs because some otherwise viable projects will not be funded owing to the principal–agent problem. In particular, those with few assets or little wealth who cannot afford to put in equity or provide collateral are more likely to be credit rationed because of the principal–agent problem.

Question 9.14 Choose the correct answer(s)

Read the following statements comparing principal–agent relationships in labour markets and credit markets, and choose the correct option(s).

  • In both cases, the principal would like the agent to exert a good amount of effort, but this cannot be written explicitly into the contracts.
  • In both cases, the duration of the relationship cannot be set as a specific time period in the written contract.
  • The problems associated with principal–agent relationships in these markets lead to some workers being excluded from employment and some borrowers being excluded from receiving credit.
  • In labour markets, one possible solution is for employers to require collateral from their employees.
  • The employer would like their worker to work hard, and the lender would like the borrower to exert effort in the project the funds were borrowed for. However, formal contracts cannot capture these elements.
  • The duration of employment is often not specified in employment contracts, but the date of repayment is a standard part of borrowing contracts.
  • One consequence in labour markets is unemployment. One consequence in credit markets is credit constraints (and credit exclusion).
  • The use of collateral can provide a possible solution in credit markets but it does not apply in the same way for labour markets.

Exercise 9.10 The rise of subprime auto loans

On Christmas Day in 2014, the New York Times published an article entitled ‘Rise in Loans Linked to Cars is Hurting the Poor’.

Here are two quotes from the article:

‘The lenders argue that they are providing a source of credit for people who cannot obtain less-expensive loans from banks. The high interest rates, the lenders say, are necessary to offset the risk that borrowers will stop paying their bills.’

‘And because many lenders make the loan based on an assessment of a used car’s resale value, not on a borrower’s ability to repay that money, many people find that they are struggling to keep up almost as soon as they drive off with the cash.’

Based on the information in the article:

  1. Who are the principals and the agents in this story? What are the loans being used for and what is the role of cars in the relationship between the principals and the agents?
  2. Using the model of intertemporal choice, explain why a borrower would take a loan like those described.
  3. Discuss the issues with payday lenders such as TitleMax, referring to the article and the quotes above.

Exercise 9.11 Microfinance and lending to poor people

Read the paper ‘The Microfinance Promise’.6

The Grameen Bank in Bangladesh makes loans available to groups of individuals who together apply for individual loans, under the condition that the loans to the group members will be renewed in the future if (but only if) each member has repaid the loan on schedule.

  1. Explain how you think such an arrangement would affect the borrower’s decision about what to spend the money on, and how hard they will work to make sure that repayment is possible.
  2. Use the concepts in this section to explain how the Grameen Bank’s lending method could affect credit rationing and credit exclusion.
  3. Find evidence about whether or not microfinance has been effective in increasing investment by groups who would normally be excluded from the credit market.

Exercise 9.12 Limits on lending

Many countries have policies that limit how much interest a moneylender can charge on a loan.

  1. Explain whether or not you think these limits are a good idea.
  2. Who benefits from the laws and who loses?
  3. Discuss the likely long-term effects of such laws.
  4. Contrast this approach to helping poor people gain access to loans with the Grameen Bank in Exercise 9.11.

Exercise 9.13 Pawnshops as a source of credit

Pawnbroking is one of the oldest sources of credit in the world. A pawnshop offers loans in exchange for items such as jewellery, which are held by the shop until the loan is repaid. Such shops are mainly used by people on low incomes. In Texas, the maximum interest rate that can be charged is 20% per month. According to a study of pawnshops in Texas, default rates are lower when items of sentimental value, such as rings, rather than items of equivalent resale value, such as TV sets, are held by the shop.

(If you want to find out more about pawnbroking, this quote is taken from Susan Payne Carter and Paige Marta Skiba. 2012. ‘Pawnshops, Behavioral Economics, and Self-Regulation’. Review of Banking and Financial Law 32 (1): pp. 193–220.)

  1. Draw a diagram with ‘Consumption now’ on the horizontal axis and ‘Consumption later’ on the vertical axis. Draw a feasible set, endowment, and indifference curves on this diagram, and use the diagram to explain why someone might choose to use a pawnshop.
  2. Define the term that is used for the items held by the pawnshop (such as jewellery), and explain its role in the pawnbroking business.
  3. Suggest an explanation for the differential default rates on the type of item held by the pawnshops when making a loan.
  1. David McKenzie. 2017. ‘Identifying and spurring high-growth entrepreneurship: Experimental evidence from a business plan competition’. American Economic Review 107 (8): pp. 2278–2307. 

  2. Suresh de Mel, David McKenzie, and Christopher Woodruff, 2008. Returns to capital in microenterprises: evidence from a field experiment. The Quarterly Journal of Economics, 123(4), pp. 1329–1372. 

  3. David G. Blanchflower and Andrew J. Oswald. 1998. ‘What Makes an Entrepreneur?’. Journal of Labor Economics 16 (1): pp. 26–60. 

  4. Douglas Holtz-Eakin, David Joulfaian, and Harvey S. Rosen, 1994. ‘Sticking it out: Entrepreneurial survival and liquidity constraints’. Journal of Political Economy, 102(1), pp. 53–75. 

  5. David B. Gross, and Nicholas S. Souleles, 2002. Do liquidity constraints and interest rates matter for consumer behavior? Evidence from credit card data. The Quarterly Journal of Economics, 117(1), pp. 149–185. 

  6. Jonathan Morduch. 1999. ‘The Microfinance Promise’Journal of Economic Literature 37 (4) (December): pp. 1569–1614.