Unit 9 Lenders and borrowers and differences in wealth

9.11 How good is the model?

A good model, as explained in Unit 2, includes what is important and leaves out what does not matter for the questions we are interested in answering.

We began this unit with some facts about wealth and credit markets and added others as the unit progressed:

• Almost everyone engages in some kind of lending or borrowing.
• In both Chambar, Pakistan and New York City, people with limited wealth often borrow at high rates of interest.
• The kinds of assets that wealthy and less wealthy people own differ: wealthy people own almost all of the shares and businesses in the US, while the main assets owned by people with limited wealth are homes and vehicles.
• Despite their need to borrow, poor people hold very little of the US economy’s total debt, except in the form of student loans.
• Many people in the US, especially those with lower incomes, are limited in how much they can borrow.

The US economy is not unusual in this respect (other than the importance of student loans).

The intertemporal choice model (Marco and Julia) and the principal–agent model (lender and borrower) together provide an explanation of these facts. Credit markets open up opportunities for both borrowers and lenders to be better off than they would otherwise be. But those with limited wealth are unable to put up collateral or equity and as a result are often refused loans or pay high rates. An exception to this is borrowing to purchase a home: people with limited wealth are often able to borrow in this case, because the home itself can be the collateral. (The same is true of vehicles.) This explains why homes and vehicles constitute virtually all of the assets held by all but the wealthiest quarter of the population. The model helps us understand these important aspects of the economy.

Like any good model, these leave out some aspects of borrowing and lending, which are important. Three of these important but omitted aspects that we will now explain are as follows:

• We are inconsistent in the way that we are impatient.
• Many forms of borrowing and lending are highly complex, and few people understand the choices that they are making.
• Most people dislike taking risks, which means that they would prefer getting a hundred dollars for certain rather than flipping a coin and getting either zero or $200.

People are not consistent: ‘My diet starts tomorrow!’

The intertemporal choice model takes account of the fact that we are impatient; but it does this in a special way. Think about something you enjoy, like your favourite meal complete with a rich dessert. The model assumes that the impatience that you feel in wanting to sit down and start eating right now compared to having the same thing, say, a day from now, is no different from the impatience you would feel about having the meal tomorrow as opposed to the day after tomorrow.

But in many of our choices, we do not act this way; the subtitle of this section about postponing your diet gives an example. Another is that people typically have difficulty in regularly saving as much for old age as they wish they had done by the time they reach the older age. The fundamental problem is that the actions we take are now, and ‘now’ is not just some date like tomorrow or the day after tomorrow, but instead has some special psychological status.

People are aware of this ‘my diet starts tomorrow’ problem (in economics it goes by the technical name, hyperbolic discounting), and we take steps to counter it. In many parts of the world, people create what are called rotating savings and credit associations (ROSCAs) in which they pool their income on payday or harvest time and set the funds aside to finance credit when needed and to avoid it being consumed in the present.

People wishing to exercise more will purchase a ten-sessions or one-month membership in a gym as a way of committing themselves to go more often (if not today). Gyms understand that people want to take advantage of this so-called commitment device, and they profit when (predictably) it turns out that buying the membership was easier than actually going to the gym and working out regularly. This raises a second problem with the model.

People are not fully informed about the results that their choices will bring about

Many of the decisions we make—like which groceries to buy—are simple in that we experience the consequences of our choices (even if these are affected by the seller’s marketing tactics) more or less immediately, and we repeat the choices often and so can alter what we buy in light of new information. But some important decisions do not have the feature of immediate feedback and opportunity to correct errors. Deciding which subject to study at university or buying a house are examples; in cases like this, we often lack important pieces of necessary information.

John Campbell, a financial economist, has identified five aspects of what he terms ‘financial ignorance’. These are ignorance of:

• financial concepts, such as how compound interest works, or what a present value is
• contract terms, namely what is being agreed to in highly complex contracts that are sometimes not even fully understood by the experts who have written them
• financial history, or how various stocks or other financial assets have performed in the past
• one’s own behaviour, illustrated by purchasing but not using the six-month gym membership
• the strategic behaviour of other actors, illustrated by the gym counting on you not to show up as often as you had hoped you would.

A result is that decisions are made that people later regret. This is less often the case for people who have substantial wealth, because they are advised by experts—and moreover, having substantial sums at stake—they have a strong incentive to inform themselves about financial matters. A result is that in many respects, financial markets do not work very well, and this especially the case in serving less well off people.

People prefer outcomes that are certain over taking risks

A third important aspect of our behaviour that is left out of the Julia and Marco intertemporal choice model is risk. We do not and cannot know the outcomes of many of the choices we make—especially where the consequences of our decisions will occur many years from now—again, choosing what to study or buying a house are examples. Marco and Julia did not worry about risk, because we assumed that any loans would be repaid, and that investments had a known and certain rate of return.

risk aversion, risk-averse

A risk-averse person has a preference for certainty (for example, getting $100 for sure) over a risky outcome of the same average value (such as a 50-50 chance of getting $200 or nothing).

Most people prefer a certain outcome to a risky bet—a payment for sure of $100 rather than a 50-50 chance of getting either $200 or nothing. This characteristic is called risk aversion: a risk-averse person prefers the sure thing. A person who is indifferent between the two is called risk-neutral.

There are many reasons why a person might be more or less risk-averse. Empirically, on average some groups of individuals are more risk-averse than others, for example, women compared to men. Like impatience, risk aversion may differ among people for intrinsic reasons. For example, someone who displays diminishing marginal utility of income might feel that $100 is sufficient and they do not care much about having more, such that having $200 is not much better than just $100. That person would then certainly take the $100 for sure rather than a coin flip for zero or $200. By contrast, a person for whom $200 really is twice as good as $100 (no diminishing marginal utility) would be risk-neutral, and would be more willing to participate in the coin flip.

As well as individual differences, the situation can also affect the extent to which a person will act in a risk-averse way. To understand this, suppose dollar values in the example above refer to the money available to purchase food over the course of a week and $100 is the minimum you could spend without risking your health. Then most people would be risk-averse: they would go for the sure $100 rather than having a 50-50 chance of starving.

The example suggests that people without wealth are likely to be more risk-averse than they would be if they had substantial wealth. This is a situational reason for risk aversion. We also would expect those who have access to credit to be less risk-averse (if your 50-50 outcome was zero, you would be able to borrow to purchase food). This is a further reason why people with limited wealth are likely to be more risk-averse: they are less likely to have access to credit.

As in the case of limited information (or financial market ignorance) risk aversion works against those with limited wealth, and can perpetuate inequalities in wealth.