President Bush spoke today about the financial crisis to the U.S. Chamber of Commerce.

I’m going to use the President’s speech as an opportunity to explain to a non-financial audience what the Federal government did this week and why. I will oversimplify in many cases, and will gloss over many details. I don’t claim that the description below is comprehensive. But it is, we think, a good starting point for discussion.

This is a story that evolves over time as we learn more, and will be debated by economists and historians long after we’re gone.

  • We begin with a global credit boom. A dramatic increase in worldwide saving outside the United States, and especially in Asia and the Middle East, meant there was a lot of money to lend. Fed Chairman Ben Bernanke referred to this as a “global savings glut.” The U.S. has a productive economy and a strong legal framework that protects investors, so a lot of this capital was attracted to the U.S. This lowered interest rates here, creating abundant and inexpensive credit. This was particularly true for riskier borrowers – as the supply of loanable funds increased, the interest rates charged to these borrowers came down a lot, making it easier for them to get loans. In many cases this was a good thing – many low-income people who had previously been unable to buy a home were able to do so. At the same time, an economist would say that “credit spreads narrowed dramatically,” and many would say this led to an underpricing of risk. Lots of lenders seeking higher yields made increasingly risky investments.While most of the focus has been on housing, and I’ll use housing to explain the rest of the story, the underpricing of risk existed in other markets as well (e.g., commercial real estate). Also note that the credit boom was not confined to the U.S. – Australia, the U.K., France, and Spain also experienced housing or credit booms.
  • We then look at a domestic housing boom. Cheap credit and low interest rates contributed to a building boom, soaring housing prices, and ultimately an excess supply of housing. Normally you’d expect about 1.6 million homes to be built each year. At the peak of this boom, about 2 1/2 million houses were being built each year. At a normal time, there’s about a 5 1/2 month supply of unsold inventory of homes; now there’s about a 10 month supply. When there’s excess supply, prices drop and construction of new homes plummets. This last factor meant that the “residential construction” component of GDP was shrinking, and caused an overall drag to economic growth beginning in early 2006.
  • Risky mortgages proliferated. Low interest rates combined with relaxed lending standards, a new model of mortgage origination, and innovations in mortgage products to dramatically expand the number of Americans who could get mortgages and buy homes. At the same time, these factors also expanded the universe of people who purchased mortgages and homes they could not afford.In an imperfect lending system, you’re always trading off between helping too few deserving people, and too many really bad risks who will never be able to pay off their loans. The expansion of credit and innovation in mortgage markets moved the pendulum toward a lot more people being able to borrow and buy homes than had previously occurred. Many of these people who previously would not have been offered credit are now living in their homes, paying their mortgage every month. This is a good thing. At the same time, these changes allowed others to purchase mortgages and homes that they could not afford.
  • You can try to minimize this tradeoff by doing things like fixing the lending disclosure rules, and changing requirements on lenders to make sure that a borrower will be able to afford the highest interest rate of the mortgage, and not just the teaser rate. But even after you’ve made these kinds of fixes (which the Fed did late in 2007), there will still always be a tradeoff and a value choice to make: do you want to help more higher-risk low income people own homes, at the cost of having more defaults and more bad lenders and borrowers abusing the system? Or do you want fewer abuses of the system, at the cost of fewer responsible low-income and high-risk people owning homes?
  • We then move to the secondary market for mortgages. Mortgages were bundled, guaranteed, securitized, and sold to financial institutions (especially banks). DETOUR: What is a mortgage-backed security?You get a mortgage from Bob’s Bank. You will make monthly mortgage payments to Bob’s Bank for the next 30 years. 99 of your neighbors get similar mortgages from Bob. Bob then sells the 100 mortgages to the company Fannie Mae (or Freddie Mac, or a fully private securitization firm). Fannie collects a fee from Bob and slaps a guarantee onto each mortgage – if you default, Fannie will pay the rest of the mortgage due to Bob’s Bank, or whoever owns it.Now imagine that each of your monthly mortgage payments is a pancake, and so your mortgage is a big vertical stack of 360 payments/pancakes (30 year mortgage X 12 monthly payments per year). Fannie Mae lines up the 100 stacks of pancakes/payments side-by-side, and then takes a slice of the pancake/payments stacks (e.g., the bottom pancake/payment from each stack, or in the usual case with Fannie Mae, a vertical slice of each stack). That slice is a mortgage-backed security (MBS) that consists of a portion of the payments from all 100 mortgages. Fannie Mae then sells the MBS slices back to Bob, after charging him a fee for the service. Bob then sells those mortgage-backed securities to investors for cash, which he can turn around and use to offer new mortgages to other homebuyers.Fannie Mae and Freddie Mac did the bulk of this guarantee and securitization business, while other firms securitized lower quality subprime and Alt-A loans. A deeper analysis could explain how this securitization contributed to problems in these secondary markets. Creative financial engineers further sliced and diced these mortgage-backed securities, breaking risk apart into little pieces and combining them in interesting, creative, and almost completely incomprehensible ways. (Imagine flipping and swapping some pancakes around before slicing them and you’ll have a feel for it).
  • Many banks and other large financial institutions, including some insurance companies, and Fannie Mae and Freddie Mac themselves, bought and held these mortgage-backed and other complex securities. These investors all made the same incorrect assumption: they assumed that anything mortgage-related would be safe and yield a good return. While most mortgages are safe investments, investors did not correctly understand that some of these assets were quite risky. They didn’t really know what they were buying for two reasons: (1) the underlying information about some of the mortgages was bad, because some of the loans were based on poor information (e.g. “no documentation loans”) or were made to people who were higher credit risks; and (2) many buyers of complex securities did not understand how the sophisticated financial engineering affected the risks built into these securities. In some cases, investors relied on the Fannie/Freddie brand name and didn’t do their own homework. Others relied solely on credit rating agencies that later turned out to be wrong in their risk assessments.
  • Banks and other financial institutions that bought mortgage-backed securities (and other mortgage-related investments) lost a lot of money when these securities later declined in value. Because many of these institutions (especially large investment banks) were highly leveraged, they faced a greater risk of failure from a bad bet, and the consequences of that failure were much greater. Many of those banks that did not fail still lost a lot of their capital. Some of these large financial institutions were so big and so interconnected with other institutions, that their failure would create a domino effect. This is what we call “too big to fail,” which should more precisely be called “too big and interconnected to fail suddenly.”

Example of low leverage

If an investment bank has $10 of capital and makes $50 of loans, it is leveraged 5 to 1. (The other $40 to lend comes from deposits or borrowing.) If that $50 of loans loses 10% of its value and pays back only $45, then the bank has lost $5, which is half of its $10 of capital.

Example of high leverage

The same bank with $10 of capital makes $200 of loans, and is leveraged 20 to 1. If that $200 of loans loses 10% of its value and pays only $180, then the bank has lost $20. All of its capital is gone (the bank is bankrupt), and the bank is $10 in the hole. Because this bank was highly leveraged, it took on greater risk of failure.

The major investment banks were levered 25 to 1. That’s like buying a house with only 4% down – if your home price declines by 5%, you’re “underwater.” And since many of these large financial firms relied on short-term financing to run their operations, when lenders started to get nervous and pull back from their short-term loans to these large firms, things rapidly spiraled downward.

That story gets us up to the point of a large bank (we’ll call it Big Bank) ending up in a bad position in two ways:

  1. Big Bank lost a lot of capital because the mortgage-backed securities (MBS) it bought have declined tremendously in value.
  2. Big Bank is still holding the MBS on their balance sheet. Nobody wants to buy these MBS. And if housing prices or market conditions get even worse than expected, those MBS will decline in value even more. So Big Bank has a security that is illiquid and contains the downside risk of further losses.

Big Bank’s problems show up in any combination of three different ways:

  • Capital – Because Big Bank has too little capital, they can’t lend as much. This hurts everyone in the economy who needs to borrow – students who want student loans, drivers who want car loans, small business owners who need credit to operate and to expand, farmers who borrow for seed and fertilizer, and others.
  • Liquidity – Banks normally loan money to each other for short periods of time. But now Large Bank doesn’t want to lend to Big Bank, because Large Bank fears Big Bank might be insolvent and not be around to pay them back. So Large Bank charges Big Bank more (a higher interest rate) for this short-term borrowing. We have seen this in dramatic fashion as the interest rate that banks charge either, called the London Interbank Offerer Rate (LIBOR) has spiked up. Large Bank may shorten the term of their lending – being willing to loan to Big Bank overnight, but not for 30 days. In an extreme case, Large Bank may not lend at all to Big Bank. To oversimplify, banks don’t trust each other enough to lend. This breakdown in trust/confidence among large financial institutions is a core problem in our financial system.
  • Solvency – At the extreme, a bank could lose so much capital that it is clearly insolvent. In less severe cases, either depositors or lenders to that bank might lose confidence that the bank was viable. Depositors might withdraw their funds from the bank, or lenders to that bank might stop lending. Either one of these could cause a “run on the bank” that could ultimately force the bank to shut down.


Many banks and other financial institutions, and especially big ones, lost a lot of money on bad mortgage-related investments. This caused them to lose a lot of their capital, and in many cases some of their assets are illiquid and pose additional downside risk to their balance sheets. This hurts those banks’ ability to lend, it hurts their ability to remain liquid and borrow short-term cash from other banks, and in extreme cases it can lead to a run on the bank (of depositors, lenders, or both) and insolvency.

I am indebted to Eddie Lazear and Donald Marron of our Council of Economic Advisers for their help with this note. All mistakes are my own.