Buffett is Wrong about Government Borrowing

Warren Buffett wrote an article for Tuesday’s New York Times, in which he speculated about how the federal government is going to finance its huge deficit. His conclusion? Private sector sources won’t provide enough money to finance the deficit, so the Federal Reserve will have to step in and print money. I never thought I’d see the day, but Buffett is way off base here.

Here’s a summary of Buffett’s analysis: The federal deficit is projected to hit $1.8 trillion this year. Buffett estimates that around $400 billion of that will come from foreigners, and that another $500 billion will be made available through the saving activities of American citizens. That leaves a shortfall of at least $900 billion, which Buffett claims must be provided by the Federal Reserve in the form of freshly printed money.

This analysis is wrong on several levels. For starters, it appears that Buffett considered only one source of domestic savings, namely households, and neglected saving activities undertaken by businesses. More importantly, he ignored private investment, which draws from the same savings pool as federal borrowing. Buffett actually alluded to this fact in his article, but suggested that if private investment were factored into the analysis, it would make the savings shortfall even worse. On the contrary, the problem largely disappears when private investment is included.

The discussion will be more illuminating if we focus on changes in saving and investment, rather than absolute levels. I’ll use 2007 as a base year (the recession didn’t start until December of that year), and discuss changes between then and the first quarter of 2009 (i.e., the latest quarter for which I have data). Take a look at the table, below, which shows Buffett’s numbers for the two periods.

These figures come from the Saving and Investment table of the National Income and Product Accounts, provided by the Bureau of Economic Analysis. The Q1 2009 figures are annualized. They are different from Buffett’s more recent estimates, but the differences won’t affect my conclusions.

Household savings have increased significantly since 2007, but the increase was more than offset by a reduction in savings provided by foreigners. Meanwhile, the federal deficit increased by more than $900 billion. According to Buffett’s logic, this implies a savings shortfall of roughly $1 trillion, which the Federal Reserve must have covered by printing money.

The first problem with this analysis is that it neglects business savings. That’s an important oversight, because businesses actually save more money than households. The second problem is that Buffett’s calculation ignores private investment, which must be subtracted from private savings in order to determine how much money is leftover to finance the deficit. Take a look at the table, below, which shows gross business saving and gross private investment.

These figures were obtained from the same NIPA table as before. Gross private investment includes residential construction, which is primarily a household investment, not a business investment. Unfortunately, the NIPA tables don’t provide separate estimates for household and business investment. We’re mainly interested in the consolidated figure anyway.

There has been a huge decline in private investment. The causes are easy to identify. Residential construction has been falling for several years, following the bursting of the housing bubble. Business investment remained strong until the financial crisis began, but has fallen steeply since then, as the outlook for profits has dimmed. The result has been a $703 billion contraction in private investment spending. Throw in a $60 billion increase in business saving, and we’ve covered 75% of the savings shortfall identified in the first table above.

This is not a numerical coincidence. The economic definitions of ‘saving’ and ‘investment’ imply that the two are always equal when aggregated at the national level. I neglected several smaller entries in the NIPA table (such as borrowing by state and local governments), otherwise we would have found that the total savings shortfall was zero (within statistical bounds). In other words, there is no mystery about how the government is going to finance its additional deficit this year. The extra savings will mainly come from domestic households and businesses, both of which are scaling back expenditures and paying down debt in response to the recession.

There’s a deeper lesson to be found here. Buffett wonders where the additional savings will come from to finance the government’s extraordinary deficits. The question itself indicates a lack of understanding about how the economy works. The right question is, how much money does the government need to spend in order to soak up the extraordinary amount of savings that the private sector is currently generating?

There is no way for the public collectively to create a savings account of unused haircuts. If everyone decides to scale back on haircuts, barbers simply make less money. The same observation applies to the rest of the economy. We can’t collectively ‘save’ output (except by running a current account surplus, which doesn’t appear to be in our immediate future). If everyone decides to spend less money, the end result is that everyone works less. That’s the definition of a recession.

The budget deficit didn’t just happen to come along at the same time as households and businesses were ramping up their savings. The government is ramping up spending in order to offset private sector spending cutbacks, which are the root-cause of the recession. The government needs to run higher deficits only so long as the private sector insists on spending (or investing) less than it earns.

In other words, Buffett has it backwards: The government doesn’t need the private sector to save heavily in order to finance extraordinary deficits. The private sector needs the government to spend heavily if it wants to save extraordinary amounts of money.

No Money to Build "The New American Home"

Today's Wall Street Journal has an amusing story that exemplifies the state of the Nation's housing market. Every year at its convention, the National Association of Home Builders features "The New American Home," a high-end model home that showcases the latest innovations in building technology. According a recent appraisal, this year's New American Home will be worth $3.5 million when it's finished. The developer has already spent $800,000 of its own money on construction, and needs another $1.7 million to complete the project. Unfortunately, it hasn't been able to get a loan. At least six banks have declined to fund the project. I'm sure that the New American Home is intended to be inspiring, but at this point it's in danger of becoming a high-profile embarrassment.

Encouraging News from the Mortgage Market

The Federal Reserve just released the results of its second quarter survey of senior loan officers. The Fed uses the survey to assess the supply of, and demand for, loans to businesses and households.

During the second quarter of 2009, domestic banks continued to tighten lending standards for prime residential real estate loans. That continues a trend that's been going on for the past three years. The good news is that for the second consecutive quarter, banks reported increased demand from prime borrowers for residential mortgages. The increase in demand is probably indicative of the passing of the financial crisis, rather than the beginning of a sustainable trend. Nonetheless, it's another sign that the housing market may finally be stabilizing.

Mixed Results for Residential Construction

Recent reports are showing mixed results for residential construction. The Commerce Department said today that July housing starts were down 1% compared to June. That was a disappointing result after June's 4% rise. Single family home starts were up 1.7%, but again, that was well below the 14% increase recorded in June. Whether you focus on all housing types or single family homes, July's results don't show any significant improvement compared to June.

Yesterday, the National Association of Home Builders released the August edition of its Housing Market Index. The HMI rose by one point compared to July, and now stands at 18. That's the best result since June, 2008, but it's well below anything that could be considered 'normal'. The average level of the HMI over it's 24-year history is 52. Until the current crisis began, it had fallen below 25 only once, for a two-month period during the 1990-91 recession.

The HMI is a confidence index, based on surveys of residential builders. August's result indicates that the outlook among builders is improving, but remains subdued. That's consistent with what most economists are saying about the broad economy.

By the way, it's curious that the 'buyer traffic' component of the HMI came in at 16, while the 'expectations' component came in at 30. If buyer traffic is so light, why are builders expecting the market for new homes to improve anytime soon?

Housing Market Recovery May Take Years

There seems to be a widely held assumption that home prices will bounce back quickly. That's silly. The housing market just went through a bubble, meaning that prices reached unsustainably high levels. So why should prices return to those same levels anytime soon?

Setting aside this logical inconsistency (or is it denial?), there is plenty of precedent for long declines in home prices. Take a look at the chart, below, which shows the Case Shiller home price index for Los Angeles.

Los Angeles prices peaked in 1990, and then fell for the next six years. The peak-to-trough decline was roughly 25%. Prices didn't return to their 1990 levels until 2000

Of course, this doesn't mean that we should expect the same course of events for Bay Area home prices. For starters, prices here have already fallen more than 25%. But it does highlight the fact that home prices tend to move in lengthy cycles. We shouldn't assume that they'll quickly adjust to the latest economic news, whether good or bad. And remember, most economists are expecting a long, grinding recovery from the current recession. Why should the housing market do any better?

Housing Market May Be Stabilizing

I can't decide if I'm skeptical or cautiously optimistic. But recent results suggest that the Nation's housing market may be stabilizing. Here's a summary:

1. Housing starts for June were up 4% (on a seasonally adjusted basis) compared to May. That's for all housing types; for single family homes, the one-month increase was 14%.

Housing starts have risen in each of the last two months. The data series are clearly volatile, however, so it's too early to announce a recovery. And if housing starts have in fact begun to recover, they’re starting from a very low level. The annualized rate of starts for June was 582,000 units. Before the crisis began, the rate had fallen below 800,000 units only once in the 45-year history of the data series.

Note: In an earlier blog entry, I mentioned that housing starts historically have been an important driver of overall economic activity. So stabilization in this sector would be encouraging news for the broader economy.

2. Sales of new single family homes for June were up 11% (on a seasonally adjusted basis) compared to May. That's the largest percentage increase in the last eight years. Sales have risen in each of the last three months, and are now 16% above their March lows.

Again, some historical perspective seems important here. The seasonally adjusted annual rate of sales for March was 332,000 units – a record low. The June rate of 384,000 units was nothing to shout about either. Before the crisis began, sales hadn’t been that slow since the 1981-82 recession.

Because sales are at such depressed levels, percentage increases can be misleading. June’s 11% improvement was an eight-year record for percentage increases, but the absolute increase in sales was a modest 38,000 units (on an annualized basis). Over the six-year period represented in the chart above, there were 14 months where sales increased by more than that.

3. Existing home prices for May were up 0.5% compared to April. That’s hardly worth mentioning on its own, but it’s the first increase in almost three years, following a cumulative decline of 32% from the peak.

The bigger news is that prices rose in 14 of the 20 markets represented by the Case Shiller indices. So May’s first hint of price stabilization is broad-based, and not simply the result of large increases in a few markets.

4. Existing home sales for June were up 3.6% (on a seasonally adjusted basis) compared to May. Monthly sales of existing homes have now increased for three months in a row.


The volatility of this series should encourage caution when extrapolating the last few months worth of data. Keep in mind, too, that until May, prices of existing homes had fallen for 32 months in a row. So while sales volumes may have stabilized, the stability has been fostered by continuous price reductions. It will be difficult to argue that the market for existing homes has stabilized until prices stabilize as well.

On the whole, it seems a little early to announce an end to the housing market downturn. Housing starts and new home sales may well have stabilized. But the tenuous stability in the market for existing homes has been paid for by continuous price reductions. There’s a good chance that demand is still falling in that segment of the market. We’ll want to see at least a few months of broad-based price stability before proclaiming that the housing market has finally bottomed.

Architecture Billings Index Falls Again

After recovering from record lows earlier in the year, the Achitecture Billings Index fell significantly in June. That's another indication that economic recovery may still be a ways off.

Source: American Institute of Architects (AIA), via Calculated Risk.