According to statistics from the US Department of Labor, there are 155 million people in the workforce. Of these, according to the Small Business Administration, roughly 45 million (30%) work for companies sized from 10 to 500 employees. Understanding these mid-sized companies is the key to understanding the current situation.
Some of the statistics are a bit out of date (2002 or 2005), but the basic facts are clear. There are about one million such companies. The total annual revenue of the companies in this segment is about 6.5 trillion dollars; this represents about 30% of the private sector by revenue. The average payroll per employee in this segment is $35,000. These companies are primarily not retail companies – they primarily sell goods and services to other companies. Based on revenue per employee figures, we can say that these companies range in size from roughly $1.5 million / year to $75 million year. The mean size is roughly 37 employees, and the mean revenue is $5.4 million dollars per year. Over 50% of private non-farm employment arises from such companies.
In general, two forms of credit are available to these companies. First, banks, financial institutions, or individuals may loan them money. In general, banks and financial institutions are the most important in this segment, but individuals serve as important lenders of last resort. We may call this kind of credit money credit, and call the professional portion bank credit. Only bank credit is measured by the Fed. Second, other enterprises and individuals deliver goods and services without being paid in advance. (Payroll is always paid after services are performed; except when companies are on the verge of insolvency, consulting services and physical goods are normally delivered 30 to 90 days in advance of payment.) This second form of finance is generally called trade credit. (People generally don't consider employee payroll credit as being part of trade credit; however, it's a strong part of the credit extended to most companies. If workers feel they're at risk by extending this credit, they'll switch jobs if they can; or their productivity will decrease substantially.) Trade credit owed by a company is called trade payables; trade debt owed to a company is called trade receivables. This credit is unsecured; the creditor is at the mercy of the debtor to pay. If the debtor pays late, or refuses to pay, the creditor normally has little recourse other than to refuse to continue to deliver.
Generally speaking, trade credit is the cheapest form of credit. It's granted based on knowledge of or relationships with a company; and the amount granted is a function of the lender's trust in the borrower, and on the lender's perceived ability to make good on the lender's other obligations.
Trade credit is normally about 60 days. This means that the buying company normally pays the supplying company within sixty days of receiving goods or services. The supplier typically pays its trade creditors as slowly as, or more slowly than it receives payment. But the supplier's non-trade obligations become due much earlier. Payroll, taxes, insurance payments, and so forth all need to be paid on fixed schedules. The supplier needs to come up with this money in order to keep operating, while it waits for the buyer to pay. Creating new goods and services therefore requires capital (cash) to create new goods and services and to cover current obligations, while waiting for payment from goods and services already created and delivered. This capital is called working capital.
Most businesses are somewhat seasonal, in that sales and receipts vary over the year. In addition, random variation will cause payments to arrive at varying rates. The peak working capital requirement may be much greater than the average working capital requirement.
Banks help solve this problem by providing receivables financing. In this case, banks allow companies to convert a fraction of receivables into cash. More precisely, the company is allowed to borrow an amount determined by the company's currently outstanding receivables. Normally, a company can expect to be able to finance 60% to 80% of receivables, up to a certain absolute maximum. (For example, a company may be able to borrow up to 75% of receivables, up to a maximum of $10 million dollars. We'll come back to this limit in a moment.) Often these agreements are combined with a requirement for reducing the credit balance to zero 30 days a year. This requirement means that the company must retain sufficient cash to allow it to pay off the loan. By this, the bank ensures that the borrowing company does not commit the short-term receivables loans to long-term projects. (A cynic might say that the bank thereby ensures that the borrower doesn't actually need the money. Depending on how you define "need", the cynic might be right.)
The amount of trade credit required by a company is, of course, directly determined by how much the company must spend in order to deliver products. We can arrive at this indirectly.
In the US economy, based on IRS data, cost of goods sold (COGS) is about 2/3 of total receipts. In the mid economy, payroll is about 1/5 of receipts. Non-payroll COGS can therefore be estimated at being around half of receipts. Non-payroll COGS is normally purchased from some other company, and so we can estimate that about 50% of overall activity is a trade payable of some kind. All trade payables are some other company's receivables. Therefore, if we ignore foreign trade (generally negligible for companies in the mid-size segment), we can make a rough estimate that midrange companies finance about 1/6 * 50% * $6 trillion, or about $500 billion at any given time [1/6 comes from 60 days, two months, 2/12 of the year].
(We could refine this by looking at inventories, which also are financed, but there are compensations – the retail sector holds inventory but doesn't extend much trade credit; there are plenty of statistics to validate this level of credit.)
On the other hand, we also know that banks set the limit of receivables credit based on the cash flow generated by the company. Since receivable financing are not intended to be used as permanent debts, they must be paid off; therefore, the company must divert free cash flow to paying down the debt. Companies that are succeeding generate about 15% free cash flow. This is especially true in the mid-economy, where volumes are lower and revenues are therefore less secure. Banks use many rules, but practical experience suggests that they are unwilling to finance more than half of the free cash flow, or about 7% of receipts. So, looking at outstanding receivables, 80% of $500 billion is about $400 billion. Looking at free cash flow, 7% of $6 trillion is about $420 billion. With this cross-check, we arrive at the estimate that at least $400 billion of bank credit is normally available in the mid market.
(We can cross-check the 7% of receipts by checking H8 data. Historical H8 data for 2006 indicates that banks had about $1,125 billion of credit outstanding to companies, averaged over the year. Bank credit is normally based on retrospective data; for 2005, IRS data indicates that companies had receipts of $21,800 billion. That translates to about 5% of receipts. If we allow for the fact that the largest companies prefer to issue commercial paper, meaning that the largest companies won't normally borrow nearly 5% of receipts, then the 7% rule of thumb seems defensible.)
The amount of trade credit in use is driven by the largest debtors, who are typically large manufacturers. If they are experiencing difficulties, they delay payments to their suppliers. These suppliers, however, have to continue to operate. They must obtain cash from other sources, either by getting payment from other companies who are not experiencing difficulty, by obtaining additional receivables financing, or by reducing ongoing costs.
In 2008, mid-range companies have had unusual problems obtaining credit. Although Fed H-8 data indicates that banks are continuing to extend commercial credit, it appears that (apart from the October crisis) banks maintained their lending by switching their focus to larger organizations. This is natural. Risk analysis is labor intensive, and banks have finite staffs in their risk analysis departments. Larger customers are normally more profitable customers for banks, because of fees on transactions; and the effort needed to analyze a larger customer is not proportional to the size of the organization. Therefore, banks will tend to prefer to loan to larger companies, even in the best of times. Given that bank reserves are limited, this could mean that banks would take more profitable business from larger companies and reduce less-profitable business with smaller companies. However, large companies normally have a variety of alternatives to bank credit. Because of this, banks are normally not large sources of credit to the larger companies. They may extend conditional facilities to these companies in order to attract daily business, but large companies will obtain cheaper credit elsewhere. This changed with the credit crunch.
With the credit contraction, several things happened that squeezed the mid-economy, even though banks (in aggregate) didn't reduce credit until November 2008. Large companies lost access to other sources of credit, and drew down their pre-negotiated lines of credit. Banks can only lend up to a certain fraction of their reserves. All other things being equal, granting this additional credit to large companies forces banks to increase reserves; or else banks must reduce other credit (call in other loans) to stay within their reserve limits. Banks could not easily increase reserves by selling equity, because stock prices were falling. Bank reserve requirements were made more critical because of the deflation in the credit derivative market, and due to bad mortgage debt: banks had to write off these assets, effectively reducing reserves. Writing down assets again meant reducing offered credit, in order to stay within reserve limits. The result was that commercial credit available to mid-size companies contracted substantially. Loans were called, and credit lines were not renewed. Companies could not switch banks, because risk management resources were not available at the banks; the banks were unable to analyze the opportunity.
An earlier article ("Forget Equities Already") discussed the effect that this contraction has on mid-size companies. To summarize the argument:
- Mid-range companies are accustomed to using $400 billion in bank credit (the October post suggested $460 billion, but we might as well be conservative).
- The money corresponding to bank credit contraction (or trade credit expansion) must come from somewhere; this is all money that is put to work in the real economy.
- For most companies, the only way to meet this kind of credit requirement is to cut payroll expenses.
- Based on average employee compensation, this credit represents 11.4 million employees, or 7.3% of total employment. Another way of looking at this is that $50 billion of bank credit corresponds to 1% unemployment.
- As Sequoia Capital points out in their presentation to CEOs, when faced with falling credit and falling sales, companies must cut much more than might be suggested by a direct comparison; otherwise they end up in a death spiral. Even the fear of insufficient working capital will cause companies to cut payroll drastically.
- Company expenses cannot cause projected peak working capital to exceed what's available; otherwise the company will fail. Companies will therefore cut expenses in response to a reduction of financing (or a forced increase in trade receivables), again by cutting payroll.
- Failures and reductions in force in companies of this size will not be quickly remedied, because of the high amount of information and relationship capital represented by a single employee (relative to larger organizations): the internal damage of a layoff is disproportionately larger.
There are mitigating factors, of course. In the overall economy, about 10% of firms are in the retail trade; these don't normally deliver products on credit. These companies generate about 15% of receipts. According to SBA statistics, only 150,000 of these companies are midsize by our definition; allowing for other unusual companies with little purchased input, we might decide that only 80% of mid-size companies need average working capital.
But these mitigating factors must be balanced against the well-documented increase in credit use by the large companies. Large companies (greater than 500 employees) get 61% of all revenue; this implies that they'll normally get 61% of bank credit. According to the Fed, outstanding commercial paper at the end of 2006 was $1.9 trillion; this was larger than commercial lending at the time, which was $1.185 trillion. Commercial paper has since contracted to $1.577 trillion. If it had grown by the same amount as bank lending since that time, it would now be $2.6 trillion. This implies that large companies are potentially demanding as much as $1 trillion that they'd otherwise be financing as commercial paper. Outstanding bank commercial credit is only $1.6 trillion. This strongly implies that large companies are crowding mid-range and small companies out of the market.
Given the size of the commercial paper market, relative to commercial loans, it's clear that the largest organizations normally use almost no normal bank credit. Why would they? If they can issue more paper than banks normally issue loans, it must be that their paper is Chart 1 shows revenue distribution; allocation of bank credit will follow the same curve, all other things being equal. Chart 2 shows how commercial paper has declined, while bank credit increased.
With this data and analysis, we can proceed (tomorrow) to policy suggestions.
Chart 1. Revenue distribution by firm size
Chart 2. Outstanding Commercial Paper and Commercial Bank Credit
