Revisiting this issue two years later, we find that the two-speed economy continues to run on two tracks. Despite the improved macroeconomic picture, a deeper look shows a clear bifurcation. Large firms continue to experience a more robust recovery than small firms and entrepreneurs, and this is at least partly attributable to differing degrees of access to finance. In particular, those who are “bank-dependent” – people and businesses that rely on banks because they lack access to alternative sources of finance – are faring worse than those who can turn to the public markets.
In many cases, the data supports a direct line from post-crisis regulation to these financing constraints. In other cases, the relationship may be less obvious, but an examination of the margin of adjustment reveals a causal link. For example, some argue that liquidity in capital markets has not been affected by post-crisis regulations, citing narrow bid-ask spreads. However, a closer look at market activity since the end of the recession reveals structural changes that are leading to rising financing costs and reduced market access for smaller companies.
Similarly, surveys of mature businesses are sometimes used as proxies to assess the credit conditions facing newer firms. These surveys suggest that it’s actually weak demand – and not constrained supply – that is weighing on bank lending to small businesses. This conclusion ignores the fact that young firms and entrepreneurs rely on very different sources of finance than larger established businesses do. This is a problem because access to the types of credit that small firms rely on has been markedly reduced since the end of the recession, and the cost for those who can get access has risen. The fact that new-firm formation has been historically soft bolsters the conclusion that regulation is weighing on entrepreneurship and economic dynamism.
Recent economic indicators pointing to higher business confidence and better growth prospects among smaller firms are welcome news. They appear to reflect business owners’ expectations that the political environment is taking on a pro-business tilt, particularly given the prospects of regulatory reform. This suggests that the seven-year divergence between large and small firms may be stabilizing – but we have not yet seen evidence that the gap is closing. Ultimately, had economic activity since the crisis not been so bifurcated, we might have seen a stronger and more widespread rebound over the past seven years.
I. Slow and Uneven Economic Recovery
Despite ongoing improvements in the macroeconomic data, with gross domestic product (GDP) growth improved since the end of the financial crisis and the unemployment rate now back at pre-crisis levels, the economic recovery continues to lag previous recoveries. We estimate that if the recovery had followed the historical pattern seen since 1980, the cumulative growth in GDP since the end of the recession (mid-2009) would be more than 10 percentage points higher today (up 28%, rather than the actual 17%). A longer time horizon shows a more dramatic underperformance: the current recovery has lagged the low end of the historical range of recoveries dating as far back as the late 1940s, as Figure 1 shows.
There has been a persistent divergence between large-firm and small-firm growth rates since the end of the recession. Although this gap has narrowed recently, it is unclear whether this trend will be sustained. Two indexes of business conditions and optimism show how the recovery has failed to play out among small firms. While large firms (as shown by the Institute for Supply Management) have enjoyed a pronounced recovery, smaller firms (as shown by the National Federation of Independent Business) have lagged. This divergence, which became pronounced toward the end of the recession, has persisted and has only very recently shown signs of converging.
New-Firm Formation Has Been Weak
We see the challenging environment for small firms (those with fewer than 500 employees) reflected in the decline in the number of these businesses since the start of the crisis. Available U.S. Census Bureau data shows that the number of small firms declined over the five years that followed the onset of the crisis – the first such occurrence since the data became available in 1977 (see Figure 3).
Using a simple trend line, we estimate that if the number of small firms had grown in line with the historical pattern seen from 1977 through 2007, there would have been roughly 650,000 more small businesses in 2014. This measure of “missing” small businesses is roughly five times the largest prior gap seen in 1982. We note that 650,000 more small businesses could potentially have translated into an incremental 6.5 million associated jobs.
Weak Employment Trends at Small Firms
Large and small firms have experienced dramatically different trends in employment during the recovery. Figure 4 shows the cumulative change in employment at smaller firms (those with fewer than 500 employees) relative to larger ones (those with more than 500 employees) between 1977 and 2014. Historically, the cumulative change in employment at smaller firms had outpaced the comparable figure for larger firms. But in recent years, this trend has reversed, with the cumulative rise in employment at smaller firms running significantly below the cumulative increase at larger firms. Other data series – including small business surveys and the U.S. Bureau of Labor Statistics (BLS) firm-employment data set – suggest that there has not been a meaningful reversal of this new pattern since 2014.
Small Business Wage Growth Has Lagged
Wage data also highlights the divergent positions of small and large establishments. Indexed to 1996 levels, wage growth at establishments with more than 500 employees outpaced wage growth at smaller establishments by a cumulative 6 percentage points during the 14 years from 1996 through 2009. Over the subsequent seven years, the gap expanded by an additional 11 percentage points, roughly twice the divergence seen from 1996 through 2009 and in just half the time (see Figure 5).
Self-Employment Has Also Been Weak
Self-employment plays a key role in the economy and can act as a safety valve for unemployed workers. Given the severity of the recession, growth in this category should have been strong. Yet the recovery here has been soft, despite recent gains. Two categories of data illustrate this point. First, the U.S. Census Bureau found that the number of sole proprietors grew by just 2% on a compound basis from 2010 through 2014; this is less than half of the rate of increase seen over the comparable time frame following the end of the 2001 recession. Second, a longer-running data set from the BLS shows that there were nearly 80,000 fewer unincorporated self-employed workers in 2016 than in 2010 – despite a rebound in the past two years (see Figure 6).
II. Access to Finance Is Critical
Why have new and small firms underperformed relative to larger firms? One critical problem appears to be their limited access to bank credit. We see direct links between post-crisis regulation and restricted finance, despite official arguments to the contrary, and this trend will continue to exert a significant impact on the dynamism of the broader economy.
New regulations since the crisis have succeeded in increasing the safety and soundness of the banking system. But they have also effectively acted as a “tax” on banks, changing the relative prices of different activities, making some activities more expensive and others cheaper. The impact across bank customers is uneven: customers who can find less-expensive sources of financing turn to them, while customers without alternatives are forced to bear the higher costs of the taxed activities or are unable to access credit. As bank credit has become more expensive, small firms have been more affected by regulations than large firms with access to the public markets. Figure 7 shows that on average, spreads in most lending markets are higher now than they were before the recent recession. We compare average spreads over the relevant non-bank benchmark since 2013 – which follows the implementation of many post-crisis regulations – against the corresponding average spread from 2000 through 2007 (see the column labeled “post-crisis vs. pre-crisis”). Doing so makes it apparent that different dynamics are playing out across the consumer and corporate lending markets.
Figure 8 plots the relationship between the change in lending prices and the degree to which each financial activity has been affected by post-crisis regulation. The Y axis shows the change in spreads for each lending category compared to the pre-crisis average, while the X axis is a measure of the impact of new regulations as well as the degree to which the activity is dependent on banks for origination. We conclude that the lending markets that have been most exposed to post-crisis regulatory changes have seen rates rise most significantly, while the markets that are least exposed – where strong non-bank alternatives exist – have seen lending spreads fall from the pre-crisis period.
Federal Reserve surveys and commentary posit that reduced lending to small businesses reflects a lack of demand rather than constraints on supply. A closer look at the dynamics of small-business lending suggests otherwise. Credit remains available to established small firms, but it comes at a higher price than before the crisis. This is problematic because higher capital costs make small businesses less competitive with larger firms that have access to public financing markets at historically low interest rates. While unrated business loan spreads are nearly 150 basis points (bp) higher than they were in the pre-crisis period, large corporations can issue investment-grade debt with spreads that are in line with what they paid before 2008. Also, spreads on high-yield corporate debt issuance are lower now than they were before the crisis.
We see the impact of the higher cost in the muted recovery in bank lending to small businesses. Although outstanding commercial and industrial (C&I) loans for less than $1 million are nearly 20% above the trough seen in 2012, they remain below their peak in 2008. In contrast, larger C&I loans outstanding (those above $1 million), are more than 45% higher than the 2008 level (see Figure 9).
Regulatory commentary also often obscures the fact that the “small” firms that are able to access bank loans aren’t all that small. Consider that for a small business to be eligible for a loan guaranteed by the U.S. Small Business Administration, it must meet a range of eligibility requirements, including being deemed able to “repay the loan on time from the projected operating cash flow of the business.” The ability to repay may be questionable for a new or young business, particularly because around half of new establishments don’t survive for five years.
Firms that are too small and that lack the collateral to secure a small-business loan have typically turned to personal sources of bank funding, such as credit cards or home-equity lines of credit. Access to these sources of capital has also dwindled since the financial crisis.
Consumer credit is more expensive and less freely available, affecting young firms and individuals alike. Credit-card spreads have risen significantly for all borrowers, up 325 bp relative to pre-crisis levels for borrowers with higher FICO scores and 360 bp for those with lower scores. Access has also changed: there are 50 million fewer credit card accounts in the U.S. today than at the peak in mid-2008. We estimate that between 2009 and 2012, about 30% of lower-credit borrowers (12 million people) lost access to credit cards.
Entrepreneurs frequently use their homes as a source of collateral for a loan and the home-equity loan market illustrates dynamics similar to credit cards, with average post-crisis spreads roughly 110 bp more expensive than they were pre-crisis. In addition, this source of funding is only available to borrowers who have home equity – which is not always the case, particularly for younger entrepreneurs.
… but it’s harder to become a homeowner today because access to the mortgage market has become restricted … Mortgage lending has become increasingly restricted to higher-credit borrowers. Figure 10 compares the pre- and post-crisis distribution of all borrowers by FICO credit score. Borrowers today are clustered at the top range of the FICO scale, while those with lower scores have been virtually closed out of the market.
… and mortgage costs have risen for everyone. Spreads on jumbo mortgages are nearly 50 bp more expensive than they were before the recession, and conforming loans are nearly 20 bp more expensive. Spreads for mortgages issued by the Federal Housing Administration and the Veterans Affairs Department are nearly 45 bp more expensive than they were pre-crisis; this is particularly concerning since this category of lending supports borrowers with lower credit scores or high debt-to-income ratios. The subprime market has contracted so severely that spread differentials here have essentially become irrelevant.
Some suggest that this is not a problem, arguing that the “missing” consumer credit had gone to individuals who should not have had access to finance in the first place. Under this line of reasoning, the collapse of the subprime mortgage market and the withdrawal of credit card finance are beneficial rather than harmful to the broader economy.
Yet it is important to see the distributional consequence of these changes. The would-be borrowers who cannot obtain affordable credit today are precisely the people who need it most. They are often lower-income consumers whose only alternatives are non-bank institutions, such as payday lenders, that charge considerably higher fees.
III. Smaller Businesses Face Tighter Credit Conditions in Public Markets
Even public markets show evidence of a bifurcation between larger and (relatively) smaller firms. Public debt markets are a critical source of funding for firms that have made it past the startup phase and that seek financing at a lower cost than they can get from banks. Traditionally, these firms have turned to public debt markets. But these markets, too, have been subject to a host of regulations and market changes that have affected liquidity dynamics, with clear consequences for smaller and would-be issuers.
The debate as to whether market liquidity has declined – and if so, what role regulation has played – continues. Regulators have concluded that liquidity has not been affected, citing narrower bid-ask spreads and higher trading volumes. To the extent that they do acknowledge some changing liquidity dynamics, regulators attribute this to changes in market structure, notably the rise of electronic trading, and downplay the role of regulation. Some also argue that the lost liquidity was “false,” heightened by bubble-like conditions prior to 2008, and that we are now simply seeing a return to “normal,” preferred levels.
Changes in liquidity can be measured in many ways, and a closer look at debt-issuance activity since the start of this decade suggests that liquidity problems are playing out through higher financing costs. New issuance overall has been robust since 2010. But if we break it down by firm size, we see a differentiated picture (see Figure 11). Issuance by the largest firms has more than doubled since 2010, while smaller firms have seen a decline over the same time frame, with the drop most pronounced for the smallest issuers.
We attribute this bifurcation directly to changes in market liquidity. With brokers unable to hold inventory, the secondary market for smaller issuers’ debt has tightened. Investors prefer larger issuances – the size of the average new issuance in 2016 more than doubled since 2010. The impact is akin to that seen in bank lending: higher costs tend to reduce small firms’ ability to borrow. When lower liquidity puts public debt markets out of reach of small firms, it impedes their ability to grow.
Putting this all together, we find that regulations have raised the cost and lowered the availability of credit to individuals and to businesses that have traditionally relied on bank financing, in ways that have already had a notable impact on the overall economy. This ultimately dampens job creation and innovation.
Recent economic indicators pointing to higher confidence and better growth prospects among smaller firms are welcome news. The outlook may be indicative of expectations that the political environment is taking on a pro-business tilt, particularly given the prospects of regulatory reform. The seven-year gap between large and small firms may be stabilizing – but is unlikely to be closing. It could take years for smaller firms to make up the difference, if that is even possible. Large firms may by necessity become the drivers of economic dynamism, but this will represent a significant shift and is likely to weigh on U.S. economic activity in the future.
About the Author:
Steve Strongin is head of the Global Investment Research Division at Goldman Sachs. He is a member of the Management Committee and the Firmwide Client and Business Standards Committee. He is also co-chair of the Firmwide Technology Risk Committee. Strongin has previously served as the global head of strategy research and co-chief operating officer of Global Investment Research, as well as the global head of commodities research.
He joined Goldman Sachs in 1994, was named managing director in 1998 and partner in 2002. Prior to joining the firm, Strongin spent 12 years at the Federal Reserve Bank of Chicago, most recently as the director of Monetary Policy Research. He serves as a director on the boards of Ocean Conservancy and New York City’s Fund for Public Schools. He is also a member of the Visiting Committee to the College at the University of Chicago and the Advisory Board to the RAND Center for Corporate Ethics and Governance. Strongin earned undergraduate and graduate degrees in economics from the University of Chicago and a graduate degree in management from Northwestern University’s Kellogg School of Management.