Small businesses are key to reverse economic stagnation – solves weak job growth and labor outflows

As the U.S. economy enters its 28th consecutive quarter of expansion (4th longest since the great depression), there is ongoing debate as to whether labor markets, and for that matter, the broader economy is nearing the end of the expansion cycle. While there is evidence that expansions do not die of old age, there are signs that the labor market recovery is nearing retirement, similar to a growing share of the labor force. 1 For example, a recent report from the BLS suggested that job growth was the lowest in six and half years. Moreover, the auspicious signs of strong flows back into the labor force reversed course dramatically. While these measures can be volatile, there are many signs that cyclical recovery is nearing its peak at a moment when conditions remain below the economy’s pre-crisis potential.¶ That being said, a broad view of the labor market suggests that conditions could not be better. The economy has added an average of 226K jobs per month since 2014, and the unemployment rate now stands at 4.7%—the lowest rate in nine years. In addition, the number of people choosing to re-enter the labor force and begin work was at an all-time high in April, pushing up the labor force participation rate, a trend that had coincided with wage gains and tighter labor market conditions. ¶ To assess if there is remaining slack in the labor market or if the remaining headwinds are structural in nature, we exploit the empirical relationship between the unemployment rate and jobs vacancies, known as the Beveridge Curve (BC), and a time-series derivation of the gap in the BC referred to as the “curve shifter.” Theory suggests that higher levels of unemployment (larger pool of job seekers) is associated with a lower number of vacancies given that a higher supply of job seekers and greater demand for employees will increase the likelihood of employers finding a match for their vacant positions. As a result, traditional business cycles produce movements along the curve. However, during the recession and recovery, there has been a persistent outward shift in the curve, which is unlikely to be explained by normal cyclical forces.¶ To put this into perspective relative to the pre-crisis, for any given job vacancy rate, there was a 2pp higher unemployment rate (UR). For sectors most acutely impacted by the crisis, such as construction and transportation and utility, the gaps were 5.1pp and 2.3pp larger, respectively. Although there have been some indications of improvements, based on a derivation of this gap or shift in the BC, there appear to be remaining frictions.¶ Using this reduced form representation of the curve shifter, we confirm previously established relationships between productivity (Lubik 2012), uncertainty (Liu & Leduc 2015), and secular shifts in labor force activity a la Bova et al (2016). However, we find that access to credit, in addition to significant fiscal policy tightening, are the key elements in the breakdown of labor market matching. Moreover, we find heterogeneous impacts across firm size and age, and industries. In fact, the reduced form representation of the factors that can “shift the curve” a la Pissarides (2000) and Liu & Leduc (2015) shows that despite a handful of cyclical indicators suggesting vast improvements in the labor market, there remain significant structural forces at play.¶ Usual suspects explain remaining gap in labor market¶ After controlling for changes in unemployment benefits (extension of unemployment insurance claims) due to the financial crisis, we found that there were three key factors that explain the significant and persistent outward shift in the BC: labor force outflows of those 55 and older, cyclical fiscal policy tightening and credit availability. In fact, over a four year cycle, increased willingness of banks to lend in the commercial and industrial (C&I) and commercial real estate loans (CRE) spaces explains 35% of the shift, while cyclical fiscal policy shocks explain 31% of the movement, productivity explains 15%, and retiree outflows from the labor force explain an additional 13%. Unlike Liu & Leduc (2015), we find that policy uncertainty explains only a small portion of the shift (3.0%). On a short time horizon (four quarters), however, productivity plays a more important role in matching, explaining nearly 50% of the shift, with labor force outflows explaining an additional 17%.¶ These results imply that factors that are cyclical by nature can have lasting effects on broader labor market activity. A process sometimes referred to as hysteresis. For example, fiscal policy shocks and credit availability tend to ebb and flow with the business cycles, leading to predictable movements along the BC. However, without offsetting shocks to other determining factors, e.g. lower uncertainty or higher productivity, the impact can persist for years, and in some cases never fully recover. In addition, to the extent that credit and fiscal policy have experienced a permanent shift, and the fact that monetary policy is becoming a less effective tool, it will be hard to envisage any significant reduction in the gap in the medium-term. ¶ In addition, we tested the impact hiring rates have on BC across firm age, size and industry. In the short run, large and incumbent firms have the largest impact on the BC. However, after six quarters, the impact that startups and small firms have on the BC is more significant and persistent. In fact, at 16 quarters, the impact that startup hiring has on BC compared to old firms is 10 times greater; for small firms, in a similar vein, the impact is 250% greater at 16 quarters versus four quarters from the initial shock. With this in mind, creating an environment that encourages small business formation and risk taking could counteract headwinds plaguing the labor market. ¶ Firms at all levels are susceptible to credit cycles and cash flow volatility. However, these factors are amplified for new entrants or small firms, imparting a larger influence on the broader labor market, particularly in keystone sectors such as retail, real estate and construction. In other words, hiring slowdowns in these sectors have the largest and most persistent effect on the BC shifter. For instance, a one standard deviation drop in the hiring rates in these sectors would shift the BC outwards by approximately 10%. To put this into perspective, in the aftermath of the crisis, the BC shifter increased by 30% from peak-trough. Moreover, unlike other industries that are more apt at weathering cycles and have greater access to credit such as manufacturing, these industries are generally slow to respond and as a result, accumulate losses over a longer time period, and in some cases, never recover.¶ This finding has substantial implication for the health of the broader economy given the fact that small and new firms are the dominant force in net job creation in the U.S.; whereas, larger (500+ employees) and older (11+ years) firms are historically net job destroyers. Moreover, startups and small businesses hire at a rate between 1 to 2.5 times higher than older and larger firms. Labor-intensive service sectors such as retail also have persistently higher levels of hiring, in some cases 50-75bp higher than other sectors. More specifically, startup hiring rates in industries such as finance and insurance, manufacturing and information are 380%, 330% and 270% higher, respectively, than their industry peers. Small and young firms in arts and entertainment, agriculture and accommodation hire at rates 200%-400% higher than the national average. Research has also shown that only 3% of businesses can be classified as “high growth businesses,” but they are responsible for a disproportionate share of growth. Moreover, small businesses are essential parts of U.S. supply chains given that can lower logistical costs, are nimble problem solvers and are better equipped to partner on joint innovations. ¶ There is additional upside to targeting small businesses and startups given that as outflows from the labor market intensify, there could be adverse effects for aggregate productivity as new entrants take time to develop skills. As a result, focusing on new business hires could help to accelerate the demographic transition.