The small business sector in recent recoveries.
Business Economics (2011) 46, 214-228.
Keywords: small business, economic recovery, employment, financial crisis
Small businesses produce half of private GDP, employ half the private workforce, and are the source of most job creation in the United States, accounting for roughly two-thirds of new net private sector jobs over the past two decades.(1) The importance of small firms to production and job creation makes their health fundamental to economic growth. Data from the National Federation of Independent Business's (N FI B) 350,000 member businesses indicate that in the current recovery period, these firms are suffering from unusually low levels of sales and earnings, investing less, and hiring fewer workers than in previous recoveries, exacerbating the nation's unemployment situation and contributing to a lackluster rebound from the most severe U.S. recession since the Great Depression. Statistics on the small business sector are not readily available, but the NFIB data suggest that this sector is dramatically underperforming. Because half of the economy is not growing, we have not enjoyed the kind of growth experienced in past recoveries.
The abnormally sluggish performance of small firms in the current recovery period naturally raises the question of "Why?" One possible explanation is that residual effects of the recent financial crisis are inhibiting the ability of small firms to grow and hire. This view is consistent with research that has found that recoveries from recessions associated with financial crises tend to be slower than average. (2) Financial crises, defined broadly as episodes during which widespread disruptions to financial institutions and markets occur, have diverse causes and take various forms. These episodes include some combination of very high inflation, currency crashes and debasements, banking crises, and cases of outright-or near-default on government-issued debt. An additional complication most common to banking crises is the bursting of an asset price bubble, usually during the run-up to the actual banking crisis.
The financial crisis of 2007/8 involved the failure of two of the nation's largest investment banks, the closures of hundreds of commercial banks, the near-default of large government-sponsored enterprises, and the bursting of a housing bubble nearly a decade in the making (and a significant part of the small business sector). Conventional wisdom suggests that the practical implication of such events is to limit the ability of households and businesses to obtain credit due to a reduction in the number of financial institutions, fewer funds available for lending, and a more judicious review of lending opportunities by creditors, all of which contribute to higher interest rates and restricted credit. Financial crises featuring a collapse in asset prices have the added complication of encouraging private agents to address balance sheet problems caused by a sudden reduction in net worth. Consumers and firms, attempting to address imbalances between newly revalued assets and liabilities, typically retrench by saving more and consuming less for a time. One aspect of this retrenchment is that the decline in asset values among the wealthy is likely related to the decline in venture capital availability.
Tighter credit conditions and reductions in consumption and investment have contractionary effects that, during a recession, add additional stress to an economy already suffering from negative growth and high unemployment. Additional downward pressure on consumer spending is particularly harmful since it is the consumer and new construction that usually lead the economy out of recession. Without sufficient demand, businesses have no reason to increase production, expand, or hire new workers, and job growth suffers as a result. The impact of a sharp reduction in spending is exacerbated when preceded by a large expansion of capacity (too many retail outlets, restaurants, inventory, strip malls, as well as new homes) as was the case in this recession. This expansion of capacity is financed by a decline in the saving rate and a rapid increase in debt. Having bought "'too much stuff in the boom, consumers have plenty of discretion about spending, as illustrated by the rapid decline and slow recovery of vehicle purchases. This excess-and-retrenchment theory is a plausible explanation of why recessions associated with financial crises are typically more severe than average in duration and amplitude and why their ensuing recoveries take longer than usual. (3)
Judged by conventional metrics like output and unemployment, the recent downturn certainly ranks among the worst in U.S. history. Starting in 2008:Q1, real GDP decreased in five out of the next six quarters, shrinking nearly 4 percent. From peak to trough, private sector employment fell 6 percent (or 8.4 million workers), causing the unemployment rate to rise to 10 percent, a level not seen since the early 1980s. While the recession officially ended in the second quarter of 2009 and economic conditions have improved since then, GDP and employment growth have lagged by historical standards. In the seven quarters following 2009:Q2, real GDP increased, on average, 2.8 percent each quarter. The economy is expanding at a rate similar to how it performed following the milder 1990/1 and 2001 recessions, but severe recessions are generally associated with robust recoveries--witness the five quarters beginning in 1975:Q2 or 1983:Q1 (both NBER troughs) during which quarterly real GDP growth averaged 5.5 and 7.8 percent, respectively. Real GDP growth in the current recovery is less than half of what might be expected based on past GDP patterns. The labor market picture is no rosier. More than eight quarters after the business cycle trough, the unemployment rate as of August 2011 remained at 9.1 percent and 7 million jobs destroyed during the recession have yet to return. Clearly, the evidence at the macro level during this recession supports the view that recessions linked to financial crises are unusually severe and that ensuing recoveries are less robust than the norm.
As for the small business sector specifically, NFIB data indicate that small firm performance has been weaker throughout the current recovery than during any other recovery since 1973. Some observers argue that it was restricted credit availability precipitated by the financial crisis that caused economic activity in the small business sector to lag in this recovery. However, the NFIB surveys that have tracked small business owners in every recession recovery since 1973 offer an alternative explanation. If the credit supply explanation is to be believed, then it should be case that small firms, on average, perform better during recoveries not linked to financial crises than during recoveries that are linked. And, more directly, owners of these firms should report "financing" issues as their top business problem.
Has the small business sector historically performed worse in recoveries following financial crises? Answering this question first requires a cataloguing of U.S. financial crises and any "associated" recessions. (4) According to a popular taxonomy invented by Reinhart and Rogoff , the most recent crisis beginning in 2007 was just the second financial crisis to hit the United States since 1929, the other being the savings and loans crisis originating in the mid-1980s.5 Using this classification system as guidance, the IMF categorizes the 2008/9 downturn as the first U.S. recession associated with a financial crisis since the 1930s. (5) This view holds that financial crises are phenomena that historically have been encountered rarely in the United States and that recessions associated with financial crises have been rarer still. Alternative approaches to cataloguing financial crises exist, however, and can yield very different results. One noted expert [Mussa 2009] counts at least seven financial crises to hit the United States during the past 50 years, including ones occurring in 1980-82 and 2000-2002, which both coincided with economic recessions. (6) Still other taxonomies may give different answers, but these two methods at least provide a starting point from which a cross-recession analysis can begin.
So what does the evidence say? The answer is: It depends. Ignore for the moment the current recovery--which is considerably worse than all other recent recoveries, regardless of whether they are linked to financial crises--and consider Figure 1, which presents time paths for the NFIB Small Business Economic Trends (SBET) Optimism Index (7) (a measure of overall small firm performance) following the five most recent recessions since 1970. For this particular analysis, Reinhart and Rogoffs taxonomy is uninformative, since none of the other four recoveries is associated with a financial crisis. Mussa's cataloguing is more interesting since using his approach (and the IMF's convention) we have two occasions of economic recovery following recessions linked to a financial crisis: the recovery beginning in 1982:Q4 and the one starting in 2001:Q4. (8) In neither of these cases did the Optimism Index consistently underperform its paths during other recoveries (not associated with financial crises) beginning in 1975:Q1 or 1991:Q1, as shown by Figure 1.
Comparing the performance of the Optimism Index during the first eight quarters following each of the NBER troughs, it is clear that the Index starting in 1982:Q4 outperformed all other recoveries over eight quarters. This is to be expected, based on a "the steeper the decline, the stronger the recovery" view of the economy. Meanwhile, the Index had a "middle of the road" performance beginning in 2001:Q4. The Index actually performed worse than either of these two time paths during the initial stages of the recovery from the 1973/4 recession, despite that recession not being technically associated with a financial crisis. From Figure 1, it is also obvious that the current recovery path is considerably worse than any of the previous four recovery paths, whether those recoveries followed recessions that were linked or not linked to financial crises.
The view that economic performance lags during recoveries associated with financial crises is too simplistic. Reality frequently resists generalizations, and the NFIB data suggest that there are sectoral differences in how the economy responds to recessions linked to financial crises as well as an historical resilience in the small business sector to disruptions in financial and credit markets caused by these crises--the current recession being an exception in the case of resilience. Thus, an explanation is needed for why the performance of small firms during the current recovery is so drastically worse than in all the others. One possible answer suggested by the data is deterioration in consumer demand to a degree not witnessed in other recession/recovery periods. The percentage of small firms indicating weak sales as their foremost business problem in the current recovery far exceeds percentages reported during other recession/recovery periods. Weak sales translate into lower revenue and reduced profits, which contribute to weaker performance. Houses were not the only things built to excess in the boom; there were too many retail outlets, restaurants, and strip malls built; and too much inventory accumulated based on spending by consumers who saved little and borrowed much. When consumers decided to save, sales growth failed to meet expectations, and the reduced level of spending was spread over many firms.
A second answer potentially lies in the confidence of small business owners about future prospects. Expectations play a fundamental role in the actions and decisions that economic agents make in the present, including decisions by business owners on whether to hire new employees, invest in new equipment and capital, and expand their businesses. In the current recovery, owner expectations about improved future business conditions are decidedly lower than they were in previous recovery periods. Misgivings and uncertainty about the future that discourage owners from investing and expanding as robustly as they otherwise would are a plausible explanation for the present lack of capital investment and anemic job creation at small firms. The NFIB survey does not provide all the details behind why each owner expects business conditions, for example, to deteriorate; but the expectation of a weaker economy is definitely correlated with lower investment spending and hiring.
Policymakers have already learned a great deal from this historic period. Certainly this recovery is a strong data point that contradicts the previous trend of declining macroeconomic volatility and increasing uniformity of recessions. Prior to the 2007/8 financial crisis, many experts contributed to an extensive literature attempting to explain the noticeable decline in the variability of output and inflation since the mid-1980s, an event and period referred to as the "Great Moderation." Three general explanations are typically offered to explain this phenomenon:
1. Structural change explanations emphasizing changes in institutions, technology, business practices, or other structural features of the economy that improved its ability to absorb shocks;
2. Arguments that "preemptive" macroeconomic--and particularly monetary--policy has permitted major recessions to be exchanged for smaller, policy-induced recessions (9) and
3. A belief that "luck" in the form of smaller, more infrequent shocks hitting the economy is the main reason why the United States enjoyed an extended period of low inflation and robust growth.
Whichever of these explanations is correct (and they need not be mutually exclusive), current events have at the very least diminished beliefs that "the future of stabilization looks bright" and added a new, unhappy chapter to the story of stabilization policy, which had been until recently, in the words of Christina Romer , "one of amazing success." This recovery period also serves as a lesson on the limits of fiscal and monetary policy, both of which have been applied by policymakers to extreme degrees in an effort to temper the recession and return the economy to a positive growth trajectory.
These lessons are important, but given that the current hole in employment is in the small business sector, we believe that no understanding of this tumultuous period is complete without considering the recent dynamics of small business economic trends. Appreciating these dynamics may assist policymakers in the development of policies capable of catalyzing small firms to create jobs, thereby strengthening the nascent economic recovery. Failure to understand the trends in this important half of the economy may lead to misguided policy that, at best, does no harm and, at worst, is counterproductive, wastes resources, and prolongs a "jobless" recovery.
The remainder of this paper documents the performance of the small business sector through the current recovery and compares it to recoveries from other recessions since 1973. Owner optimism, capital spending, inventory swings, and hiring patterns will be identified to pinpoint the source of weakness to better understand what policies might be implemented to encourage employment growth. In particular, the aforementioned themes of missing demand and owner uncertainty will be explored in detail to help explain why for the small business sector during this recovery period, "this time is different."
1. Employment and Hiring Plans
The Great Recession is notable for the severity of its labor market dislocations compared with other postwar recessions. In the worst stages of the recession, the national unemployment rate peaked at a level above 10 percent, a threshold surpassed only once since WWII, during the 1980-82 recession. The recent fall in the unemployment rate to 9 percent, which by itself signals improving labor market conditions, belies continued difficulties faced by the American workforce. Part of the fall in the unemployment rate is due to workers exiting the workforce: the labor force participation rate recently fell to 63.9 percent, its lowest reading since 1984. Additionally, both the share of long-term unemployed and the duration of long-term unemployment are at post-WWII highs.
Small firms were especially hit hard by the recession and reacted by cutting costs, including those for labor. The NFIB data set documents changes in the level of employment, hiring plans, and vacancies at small businesses on a quarterly basis (monthly since 1986), providing information on labor market trends at the firm level. The reductions in small business employment during the fourth quarter of 2008 and in 2009 were the largest ever recorded in the history of the NFIB data series. (10) Firms, reacting to lower-than-expected sales and falling profits, sought to return to profitability by reducing spending on factors of production. The data also suggest that these spending cuts may also have been made with a view toward rebalancing supply in anticipation of a prolonged decrease in customer demand. Related to the dramatic reduction in employment was a "fire sale" to raise cash and eliminate a huge excess inventory produced by the sudden increase in the saving rate in 2008:Q4. This produced an unprecedented surge in price cutting that many felt signaled the start of "deflation." However, once the excess inventories were eliminated, price cutting was reversed: in 2011, prices started rising sharply.
Figure 2 shows time paths for the average quarterly change in small firm employment for the past four recovery periods starting at the business cycle trough. The net change in employment per firm hit an all-time low of--1.02 workers per firm in 2009:Q2, coinciding with the most recent trough. The previous recession low was--0.56 workers per firm in 1982:Q3, about half as severe. Although the economy stopped contracting in the second quarter of 2009, changes in small firm employment remained negative throughout 2009 and most of 2010. The streak of negative quarterly job growth paused briefly in 2010:Q4 when the net change in employment was zero, a measure indicating that firms experienced neither net gains nor net losses in employment for the period. Although employment change turned slightly positive in 2011:Q2 (reported employment change in the preceding months), any suggestions that job creation in the small business sector had finally turned the corner were short-lived, as employment change was once again negative (--0.15 worker per firm) in 2011:Q3 (the July survey reports job change in 2011:Q2).
Unemployment is a lagging indicator, and continued job losses among small firms five quarters after a new business cycle begins are not unusual. Looking at the past four recessions, only in 1983:Q3 were net employment changes positive five quarters following a cycle trough. What is noteworthy about job losses in the present recovery is their severity. In the second half of 2009, small firms shed jobs at a faster rate than during the initial stages of any of the 1982, 1991, or 2001 economic recoveries. Net employment losses at small firms averaged--0.81 workers per firm in 2009:Q3 and did not fall below--0.5 workers per firm until four quarters later. In contrast, quarterly net employment change averaged--0.17 at the start (cycle trough) of the other three recoveries and never fell below--0.27 during any of the three recovery paths. As can be seen in Figure 3, most of the volatility in job growth came from changes in layoffs. New hires demonstrated less volatility, but both indicators were clearly more extreme in this recession than in any prior downturn. Job reductions have returned to historically "normal" levels, but reports of increases remain historically low.
A second noteworthy observation is that while the rate of job creation today is roughly the same as it was at this stage in previous recoveries, a considerable amount of time elapsed before the current time path converged to the level of its peers. Four quarters passed dating from the 2009:Q2 trough before job creation "caught up" with statistics from previous recoveries. Counting backward from 2011:Q1, net employment changes per firm were at or below zero for 11 consecutive quarters. This is the second longest such streak on record. Only during the 1980-82 period was there a longer period of sustained small business job loss (14 quarters). Thus, in terms of both the scale of job separations and the duration of sustained negative job growth, the impact of this recession on small business employment has been particularly acute, consistent with expectations of economic recovery following a financial crisis.
Despite recent improvements in labor market indicators, the NFIB data argue against the return of robust job growth at small firms in the near term. Supporting this view are two measures of labor demand that are trending at noticeably lower levels than they did following previous recessions. The first indicator, net job creation plans as shown in Figure 4, measures expected future demand for labor by small firms. This recovery period witnessed the first occasion in history that this indicator remained negative for four quarters following an employment trough. Net job creation plans returned to positive territory in 2010:Q3 (where it has since remained) when more owners began planning to hire workers than shed them. However, the trajectory of planned hiring has remained below where it was in previous recoveries.
This does not augur well for the economy's quick return to a state of full employment. For at least two and a half decades (the period of time for which data exist), the small business sector has been the source of most new jobs created in the United States. Lacking evidence of major structural change that might alter this dynamic, if conditions remain static, one may expect a prolonged and gradual reduction in the rate of unemployment. More rapid improvements in employment will only be achieved by policy that delivers more customers and offers a more positive view of the future.
The second NFIB labor demand measure is the reported number of unfilled job openings at small firms (Figure 5). (11) In general, a larger number of unfilled openings indicates tighter labor markets. As with job creation plans, the number of unfilled job openings during this recovery has consistently lagged the number of unfilled openings reported in previous recoveries. Since 2009:Q2, the net percentage of firms reporting positions they were unable to fill rose gradually from 9 to 14 percent in 2011:Q2 before dipping to 12 percent in the most recent quarter. Although this measure is performing better than during the depths of the recession, it is still below any other previous recovery.
2. Expectations and Investment
Decisions about expanding, hiring, buying new equipment and accumulating new inventories all depend upon expectations about the future. Modern economic theory assumes that firms (and their owners) are rational agents who incorporate expectations of the future into decisions made in the present. In this framework, a business owner's plans to hire, make capital expenditures, or buy inventories are driven by expectations regarding future sales for his/her firm, with more positive outlooks leading to greater investment and expansion in the current period. Firms will not invest in equipment that is not expected to pay for itself (just as they will not hire workers who are not expected to earn their pay in added value to the firm).
Figure 6 shows the pattern of expected business conditions through each recovery period starting with the 1974-75 period. The recovery paths resist generalities, but the relatively less optimistic expectations in the current recovery stand out. Unlike earlier recoveries in which business conditions expectations either started out at a much higher level or began at a similar level but experienced a sharp increase at the start of the expansion, the current recovery path started low and has regularly underperformed the other paths since then. The average net percent of owners expecting improved business conditions in the most recent nine quarters is more than 32 percentage points lower than the average of the previous four recoveries. Current expectations for better business conditions also lag historical expectations at this stage in a recovery. The series average nine quarters into a recovery is +18 percent, in contrast to recent negative readings (--15 percent in 2011:Q3) that are actually lower than the +2 percent reported in 2009:Q2.
Figure 7 shows the recovery paths of the net percent of owners expecting real sales to increase over the next three months. Expectations for real sales in the current recovery generally lag those of previous recoveries and help explain the historically poor performance of spending and hiring plans. No customers implies no need to hire, build inventories, or invest in expansion. Since the trough, the net percent of owners anticipating improvements in real sales had increased gradually from--11 percent in 2009:Q2 to a postrecession high of + 13 percent in 2011:Q1, a moment in time when the gap between the current recovery path and the series average was at its narrowest point since the cycle trough, before falling a cumulative 15 percentage points in consecutive quarters to--2 percent in the most recent quarter. Whatever the cause of this deterioration in expectations, it clearly signals a reduced need to hire and invest.
Expectations of future business conditions and sales should influence owners' perceptions as to whether or not business expansion is a good idea. Given pessimism over future business conditions and sales, it is not surprising that small business owners' current outlook for expansion also lags previous recoveries (Figure 8). During the past nine quarters, the net percent of small business owners who consider it a good time to expand has not surpassed +8 percent, whereas in previous recoveries the net percent usually rose quickly to levels above + 10 percent and stayed above that threshold for at least two years. The most recent reading of +6 percent (net) is marginally higher than the +4 percent reported in 2009:Q2, some nine quarters ago. Small business owners were considerably more bullish about expansion in both 1976 and 1983 than today, despite having just exited periods of very high inflation and strong recession.
Pessimism over short-run economic prospects among owners explains the sluggish job growth observed at small firms. Not anticipating a surge in demand or improved economic conditions in the future, owners are reluctant to hire more workers. This also explains why both planned and actual capital outlays in equipment, buildings, or land during the current recovery are down compared with previous recoveries. A larger percentage of owners are failing to replace old capital or purchase new capital than in the past. Since the decision to invest depends on expected return on investment, itself a function of expected revenue and costs, poor sales and growth prospects reduce the desirability of investment.
Figures 9 and 10 show the recovery time paths for actual capital expenditures made during the last six months and planned capital expenditures during the next three to six months, respectively. The 45 percent of owners reporting a capital expenditure made in the six months previous to 2009:04 and 2010:Q3 are series lows. The average percentage of owners reporting a capital expenditure over the past 10 quarters is 47 percent, well below the series average of 59 percent as well as the average during earlier recoveries (56 percent). Meanwhile, the percent of owners planning capital expenditures during this recovery has generally stayed at or below 20 percent. In 2011:Q3, 20 percent of owners indicated they were planning on making a capital purchase. In previous recoveries, this statistic had converged to levels in the 25-35 percent range by now.
3. Past Sales/Earnings
Since small firms make up the bulk of the interface between the production sector and consumers, reports of past sales trends, earnings, capital expenditures, hiring, and inventory accumulation should illustrate differences in the interactions between producers and spending agents, primarily consumers, who send signals in the form of purchases at small firms. These signals are transmitted by business owners to producers, who adjust production depending on whether the received signals differ from their expected levels. As a consequence, hiring, capital spending, and inventory investment exhibit a similar pattern to sales, with a lag.
Figure 11 illustrates how reported sales trends have performed quite differently in the current recovery period. The net percent of owners reporting improvements in sales during the current recovery has been lower than it was during similar stages of previous recoveries every quarter. A net--8 percent of owners reported improved sales in 2011:Q3. The recent lack of demand by spending agents is reflected in the percentage of firms who report "weak sales" as their most important problem. Figure 12 shows that this indicator reached an all-time high of 33 percent in 2009:Q4. Similar spikes in negative sentiment toward sales performance were also recorded following the 1982, 1991, and 2001 recessions, but never to such a degree.
Like many of the other preceding indicators discussed, earnings during the most recent recovery period also achieved historically worst levels of performance, as shown in Figure 13. A net 47 percent of small firms reported falling earnings in 2009:Q1 (the all-time low), just before the Great Recession "officially" ended. Since then, earnings during the current recovery path have been worse relative to other recovery periods every single quarter. Business owners are rational agents and will react to a prolonged fall in sales by paring expenditures. Cutting inventory and foregoing investment are usually among the first actions taken. Reductions in labor costs are usually among the last options taken, but even the most reluctant owners were forced to freeze (or cut) wages or even reduce headcount. The readings for sales and earnings provide grist to a tale of historically low levels of customer demand that caused small business owners to make the most rapid downward adjustments in wages, prices, and inventory recorded in the history of the NFIB data series.
When consumer spending plunged in the fourth quarter of 2008 and the savings rate recovered to about 6 percent (up from 1 percent), small businesses were caught with a huge excess supply of inventory that had been accumulated to support spending by consumers unable to spell "saving." This triggered an immediate "fire sale" to liquidate excess stocks and raise cash, producing the most prolonged and widespread reduction in inventories in the 37-year NFIB survey history, as shown in Figure 14. This, to a large degree, explains the record pace of price cutting that started late in 2008 and ended in 2011:Q1 (see next section). Plans to add to stocks have lagged other recoveries as well (Figure 15) because some owners are still in the process of liquidation and will not order new stock and because many owners do not expect much improvement in the economy, even if their current stocks are in balance. (For the past year, about as many owners have indicated stocks were "too low" as reported stocks "too high," historically a very lean position.)
To offset a decline in real sales, owners cut prices to attract customers and adjust costs, including inventory liquidation real sales, owners cut prices and adjust costs, including (raising cash in the process) and reducing labor costs (both headcount and compensation). Figure 16 shows that prior to this recession, the net percent of owners reducing prices had only been negative on four separate occasions. (12) The net 24 percent of firms reporting price reductions in 2009:Q2 is a record high. The percent of firms lowering prices has decreased steadily since then, indicating that firm owners have generally concluded the "fire sale" to liquidate excess inventories and that sales volumes have improved sufficiently to permit some price improvements.
A final note on selling prices: there was a very different time path for selling prices in the 1975 recovery when the net percent of firms raising prices regularly exceeded 40 percent. The large share of firms raising prices then was a reflection of the inflationary environment of the mid-1970s. During this period, the CPI inflation rate was frequently in double-digits and never fell below 7 percent.
Figure 17 shows the net percent of firms raising worker compensation during the last three recoveries. (13) As theory would suggest, the percentage of firms increasing compensation during the current recovery is substantially below levels during either the 1991 or 2001 recoveries, ranging from 0 to +4 percent until three quarters ago when the + 5 percent threshold was broken. By comparison, the net percent never fell below + 13 percent in the three years following a trough in either of the previous two recoveries. The smaller share of firms raising compensation provides evidence that the position of employers in wage negotiations has been strengthened due to persistent levels of high unemployment, although poor sales and the inability to raise selling prices also pressure firms to control costs and deter them from offering larger raises.
6. Credit Conditions
No discussion involving financial crises would be complete without commenting on the availability of credit. It was mentioned earlier that the conventional thinking surrounding financial crises was that they are followed by tighter credit conditions, as creditors who made bad lending decisions leading up to the crisis fail or retrench and approach new opportunities with greater caution. The literature on small firm credit access following financial crises is limited, but at least one study provides evidence of potential credit tightening among small firms following a financial crisis. (14) If this view holds true, then one would expect borrowers in the recent recession/recovery period to find it harder to obtain credit. The NFIB data supporting this hypothesis are mixed.
Throughout the Great Recession and ensuing recovery, few small business owners reported "credit" as their issue of greatest concern. Not once during the current period did more than 5 percent of owners indicate at any one time that arranging financing was their most important problem. Compared with other recovery periods, the percent of owners who described financing as their most important problem is on the low end of the spectrum, as shown in Figure 18, and it is clear that when it is a problem, owners report it. Credit was a much more pressing issue for owners during the recovery beginning in 1982:Q4--a recovery that, as the reader may recall, according to one catalogue of events (Mussa's), also followed a financial crisis. The percent of owners who stated credit was their most important issue remained above 15 percent throughout the first two years of the '82 recovery. The initial stages of the recovery beginning in 1975:Q1, although not technically associated with a financial crisis, also demonstrated high levels of credit difficulties among small firm owners. (15)
Of course, simply because owners indicate that credit is relatively less important compared with other problems does not mean that it is not a problem at all. Loan availability for regularly borrowing small firms (those who borrow money at least once every three months) near the trough of this cycle reached levels of distress not experienced for two decades.(16) Figure 19 shows the net percent of regular borrowers who report getting a loan to be harder than it was in their previous attempt to get financing. "Harder" does not necessarily mean that loans were refused; it could mean, for example, that more documentation (for regulators, in many cases) or collateral was required, or that the borrower may have received less credit than requested. Throughout the financial crisis and ensuing recession, reports of loan difficulties steadily worsened, reaching a high in 2009:Q3 of +15 percent (net) of regular borrowers finding it more difficult to obtain loans. It is clear that credit is easiest at the beginning of an expansion, gradually tightening into the ensuing recession (this was less pronounced in the "dot com" years).
The trend in recent quarters has been one of gradual improvement. In 2011:Q3, a net +10 percent of owners indicated that loans were harder to obtain--an improvement from recent highs, but still above the series average of +6.4 percent. Viewed in isolation, the loan availability data indicate that credit conditions did worsen for small firms. Yet, even in the worst moments of the Great Recession, they still remained better for regular borrowers than the all-time worst conditions experienced in the mid-to-late '70s and the early 1980s.
Markets are guided by supply and demand, and the loan availability data provide some insight into the supply of credit to small firms. The NFIB data set also offers some demand side credit indicators: (a) the percent of owners who find their borrowing needs satisfied (or not) and (b) the percent of firms who are regular borrowers. The story conveyed by Figure 20 is one of a bump since 2007 in the percent of borrowers indicating their credit needs were not satisfied, and a rise in the percent of owners who did not try to borrow.
The decrease in the percent of owners borrowing, regularly is explained by a number of factors. Borrowers may not need to borrow, since the business outlook is bad and sales are down. Loans must be repaid, implying that their proceeds are spent on projects or employees that can produce enough revenue to repay the loan, which is more difficult in a recession. If it does not look like a good time to expand, then there is no reason to raise funds for new investment. Another factor is that discouraged owners may believe (rightly or wrongly) that credit conditions in the aftermath of the financial crisis tightened sufficiently that they would not be able to obtain enough credit to make trying to borrow a worthwhile endeavor. Separate NFIB studies of small D&B firms indicate that some owners assume the answer is "NO" and subsequently do not apply. (17) A third factor is that owners experiencing difficult economic conditions retrench not only by cutting labor costs and slashing inventory, but also by deleveraging. Some combination of these three factors may be operating for some firms. The increase in the percent of owners not wanting a loan through the recession (shown in Figure 21) and the decline in the percent of owners borrowing regularly (shown in Figure 22) invite these multiple hypotheses. Whatever the cause, a record number of owners did not want to borrow.
There is no doubt that credit got tighter after 2007, especially for mortgage credit, where there seemed to be no underwriting standards for mortgages or second loans. Many owners used this imaginary equity to help finance their operations. When housing prices started falling, available equity to support borrowing vanished, leaving only payment burdens. More importantly, of course, consumer spending was adversely impacted. Anecdotally, speaking at bankers' conventions around the country, community bankers were observed to report infrequently raising lending standards, reporting that they knew a good loan when they saw one but were seeing far fewer good applicants--no surprise in a recession. Large banks that had major losses from trading, not banking, did reduce lending to small firms. Very few owners reported "financing" as their top business problem from the onset of the financial crisis, especially compared with the 1980 82 period when spending declines were about as bad, according to the NIPA accounts.
When credit is a serious problem, owners report it. "Difficulties" in securing credit always rise as an expansion matures and peak during the ensuing recession. Difficulties were reported more frequently during the last recession than during any except the pre-1983 period when reports of financing problems were twice as high. However, blaming banks for prolonging the recession by refusing to lend to "creditworthy" owners appears to be based on a poor understanding of the demand side of credit markets. Firms have little need to borrow for expansion or hiring when there is little likelihood that such investments would pay off. Even with a zero interest rate, loans still have to be repaid! The level of excess reserves being held at the Federal Reserve indicates that banks have plenty of money to lend, and surely they would prefer a good business loan to earning 1/4 percent at the Federal Reserve if such loans could be made. Policies designed to induce banks to lend more invite the kind of risk-taking that created the housing bubble--lots of jobs were created by making bad loans, and it did not leave the economy in a good place. A repeat is not desirable.
This was not the first great recession in the history of the United States; the nation has been down this road before. But times change, and this time is certainly different from other recent downturns. The macroeconomic stabilization policy lauded so highly for producing the low macroeconomic volatility of the Irrational '90s and the early stages of the 2000s (an era also marked by irrational beliefs, such as the expectation that house prices would go up indefinitely) has failed, leading to historic levels of distress in the economy. Unfortunately, some things also stay the same, like the role major financial institutions play in market crashes. (18)
This paper documents the importance of the small business sector to the economy (a fact that seems to be routinely overlooked by macroeconomic policymakers) and how it has performed through the most recent recession/recovery period. Many conjectures have been offered about why economic growth in the aftermath of the Great Recession has been so anemic, but one thing is clear: although large manufacturing firms, tech firms, and the agricultural sector have performed well (including producing record corporate profits), the other half of the private economy, small businesses, has not grown and exhibits the worst recovery performance in recent history. A solid economic recovery requires the participation of the small business half of the private economy, and economic policy must recognize this. Averaging zero or negative growth for this half of the economy, despite large firm expansion, will produce a 2 percent figure for overall growth. This asymmetric recovery also explains how after two years of recovery, GDP exceeds its previous peak but seven million fewer workers are employed.
Since it was irrationality and the madness of crowds in the financial sector that led us into this mess, it stands to reason that the solution to the current economic malaise lies in rational policies that assist the real economy. The role of small businesses is to provide jobs and produce wealth as they compete to produce what it is that consumers want at the best possible price. Public policy that supports this engine of employment and output growth is long overdue. What the small business sector (and therefore the economy) needs are policymakers who truly understand the role of entrepreneurial firms in the creation of jobs, wealth, and products and services that consumers want. Competition, property rights, and minimal interference are critical to its growth.
(1) For more information on the makeup of the small business sector, see the U.S. Small Business Administration Office of Advocacy's FAQ sheet on their website at http://web.sba.gov/faqc/faqindex.cfm?areaID=24.
(2) A good overview of recessions and recoveries associated with financial crises is given in chapter 3 of [International Monetary Fund 2009].
(3) Explanations for slower growth during recoveries associated with financial crises include weak domestic demand and tight credit conditions. Aggregate demand falls in all recessions, but it tends to fall more following a financial crisis because of larger "booms" preceding the crisis (that is, credit growth and consumption as a share of GDP rose more than usual, and asset price bubbles may have formed before the crisis). As a consequence, consumption falls much more in recessions following financial crises since not only are there fewer workers earning incomes to fuel purchases, but households also feel a wealth effect from lower asset prices and must fix poor balance sheets by saving more. The tighter credit conditions are usually caused by a combination of stricter lending standards and reduced credit supply, as lenders attempt to recover from overexpansion during the "boom" phase.
(4) We adopt the IMF's convention by labeling a recession as being "associated" with a financial crisis if the recession episode starts at the same time or after the beginning of the financial crisis. See International Monetary Fund [2009, p. 112] for more information.
(5) The S&L crisis was produced by the rapid increase in interest rates that were part of the fight against inflation. Rates doubled, halving the value of mortgages held by these government - regulated specialized lending institutions. These institutions were not highly leveraged, simply bankrupt and not systemically linked to other financial institutions, making the resolution more manageable with a rapidly growing economy.
(6) The seven U.S. financial crises identified by Mussa  occurred in 1971, 1974-75. 1980-82. 1987, 1991, 2000-02. and 2007 to the present.
(7) The Optimism Index is an average of 10 variables from the NFIB data set, including measures of small business employment, inventory, sales, credit, earnings, and expectations about the economy since 1973:Q4. when the survey began. Although the index is but one metric, it captures trends in the small business economy from a multitude of angles and, as such, is a useful gauge of overall small firm performance.
(8) The dates indicating the start of economic recessions and recoveries are those provided by NBER.
(9) Two good sources for thorough explanations of the "improved monetary policy" view are Bernanke  and Romer .
(10) This article focuses on trends in small firm performance in the current recovery period. For a discussion on small firm performance during the Great Recession, please refer to Dunkelberg, Scott and Chow .
(11) The survey asks, "Do you have any job openings that you are unable to fill right now? "
(l2) 1983:1992:Q1, 2002:Q1. and 2003:Q1.
(13) Data on worker compensation were not collected before I984:Q2. Information on compensation changes during the 1975 and 1982 recoveries are therefore unavailable.
(14) In one study of "nonmonetary effects" of the banking crisis preceding the Great Depression, evidence was provided to support the theory that small firm credit became more expensive and difficult to obtain due to increased costs of credit intermediation (the process of channeling funds from savers. lenders to good borrowers). A higher cost of credit intermediation means that a borrower must now meet a higher "hurdle" rate in order (a) to be viewed as a "good" borrower by lenders and (b) for taking out a loan to be considered a good decision by the borrower. The net effect is fewer loans made. See Bernanke .
(15) The 1975 recovery was a period of high inflation. In such an environment, lenders will look to charge borrowers higher interest rates in order to compensate for the anticipated rapid deterioration in the value of money over time. For this reason, credit is often more difficult to obtain in times of high inflation. The 1982 recovery, meanwhile, followed a period of historically high interest rates and high inflation. For borrowers, high interest rates raise the break-even barrier for investment, whereas for lenders, they reduce the pool of attractive borrowers. The probability of a "match" between borrowers and lenders diminishes, and borrowers may perceive this as a worsening in credit conditions.
(16) The survey asks regular borrowers, "Are these loans easier or harder to get than they were three months ago? "
(17) See, for example, Dennis [2010, p. 13]. This report is the first (chronologically) of three NFIB studies of credit access/ use at small D&B firms in the current recovery period.
(18) According to the final report by the Financial Crisis Inquiry Commission , a "key cause" of the 2007/8 financial crisis (which was deemed "avoidable" ) was "dramatic failures of corporate governance and risk management at many systemically important financial institutions." The authors concluded that "there was a systemic breakdown in accountability and ethics" (by both lenders and borrowers). and that a "combination of excessive borrowing, risky investments, and lack of transparency put the financial system on a collision course with crisis." The culpability of major financial institutions in the most recent financial crisis is only the latest incarnation of a long-standing problem. Large financial institutions have played a role in market crashes throughout history. Readers interested in learning how one of today's largest investment banks played a role in financial crises both past and present may wish to read Cohan .
Bernanke, Ben S. 1983. "Nonmonetary Effects of the Financial Crisis in the Propagation of the Great Depression." The American Economic Review, 73(3): 257-76.
Bernanke, Ben S. 2004. "The Great Moderation;' Remarks made at the meetings of the Eastern Economic Association, Washington. DC, 20 February 2004.
Cohan. William D. 2011. Money and Power: How Goldman Sachs Came (o Rule the World. Random House.
Dennis Jr., William J. 2010. "Small Business Credit in a Deep Recession." NFIB Research Studies, 10(1): 13.
Dunkelberg, William C, Jonathan A. Scott, and Michael J. Chow. 2010. "Inflexible Wages and Prices? Evidence in the Current Recession." Business Economics, 45(2): 94 101.
Financial Crisis Inquiry Commission. 2011. The Financial Crisis Inquiry Report. United States Government Printing Office, http://www.gpo.gov/fdsys/pkg/GPO-FCIC/pdf/GPO-FCIC.pdf.
International Monetary Fund. 2009. World Economic Outlook: Crisis and Recovery. International Monetary-Fund, http://wvvw.imf.Org/external/pubs/ft/weo/2009101/.
Mussa, Michael. 2009. "Adam Smith and the Political Economy of a Modern Financial Crisis" Business Economics, 44(1): 3-16.
Reinhart. Carmen M., and Kenneth S. Rogoff. 2009. This Time is Different: Eight Centuries of Financial Folly. Princeton University Press.
Romer, Christina D. 1999. "Changes in Business Cycles: Evidence and Explanations." Journal of Economic Perspectives, 13(2): 23-44.
Romer, Christina D. 2007. "Macroeconomic Policy in the 1960s: The Causes and Consequences of a Mistaken Revolution," Lyndon B. Johnson Presidential Library, Economic History Association Annual Meeting, 7 September 2007.
This paper was awarded the NABE Award as the runner-up in the competition for the Edmund A. Mennis Contributed Paper Award for 2011.
* Michael J. Chow is a Senior Data Analyst at the National Federation of Independent Business Research Foundation. He has previously worked for the Office of Economic Policy and Analysis at the Department of Labor and for the President's Council of Economic Advisers. He received an MSc in Econometrics and Mathematical Economics from the London School of Economics and a BSE in Electrical Engineering from Princeton University.
William C. Dunkelberg is the Chief Economist of the National Federation of Independent Business and professor of economics and former Dean of the School of Business and Management of Temple University. He is also Chair of the Global Interdependence Center and Chairman of the Board of Liberty Bell Bank. He received BA, MA, and Ph.D. degrees in economics from the University of Michigan and has served on the faculties of the University of Michigan, Stanford University, and Purdue University. He also served as Study Director at the Survey Research Center, with responsibility for the Survey of Consumer Finances.
MICHAEL J. CHOW and WILLIAM C. DUNKELBERG *
|Printer friendly Cite/link Email Feedback|
|Comment:||The small business sector in recent recoveries.|
|Author:||Chow, Michael J.; Dunkelberg, William C.|
|Date:||Oct 1, 2011|
|Previous Article:||U.S. workers delaying retirement: who and why and implications for businesses.|
|Next Article:||An economic perspective on patent licensing structure and provisions.|