Pivot Lab #1 — Labor Constraints
Multi-company · mixed · mid market · Facilitator: Dr. Sarah Chen
Captured Hypotheses (12)
BLS data shows MI manufacturing hires at 3.5% vs separations at 3.3% — a near-zero net gain of 0.2%. At our 28% turnover, we are churning through workers 8x faster than the industry can replenish them. We've treated this as a 'talent shortage' when the data shows it's a retention failure we created.
Assumption: The tight labor market is the problem. We're competing for a shrinking pool and there's nothing we can do about it.
Map our internal hires/separation rate against BLS state data (3.5%/3.3%). If our separation rate exceeds the state average, the problem is internal. Implement exit interview protocol for 90 days, correlate departure reasons with wage percentile position (currently 22nd) and benefits cliff status.
Identify the % of separations attributable to compensation vs. other factors. If >50% are compensation-driven, the 'talent shortage' narrative collapses and the business case for wage internalization is proven by the company's own data.
BLS MI manufacturing: hires 3.5%, separations 3.3%, net +0.2%. Company turnover 28% — 8x the state net hire rate. At 72% market tightness with wages at the 22nd percentile ($4.40 below $26.40 median), the company is extracting labor capacity faster than the regional market can replace it. This is a measurable externality: the company's retention failure becomes the community's workforce instability.
BLS wage data shows MI manufacturing P25=$19.50, median=$26.40, P75=$35.80 — a $16.30 spread. We pay $22, clustered just above P25. For $5/hr more per worker (~$2.8M/year for 273 workers), we'd move from P22 to roughly P55. The question is whether the $2.8M buys more in retained productivity than it costs in payroll.
Assumption: A $5/hr raise across 273 workers is unaffordable. The $2.8M annual cost would destroy margins.
Model the true cost of current wages: (turnover cost × turnover rate) + (public assistance externality per worker × eligible workers) + (quality defect cost attributable to inexperience). Compare against $2.8M raise cost. Run the model using actual BLS wage band data to identify the optimal wage point where total cost of employment is minimized.
Financial model demonstrates the break-even wage point. If current turnover costs + externality costs exceed $2.8M (likely, given $45K per separation × 76 separations/yr = $3.4M), the raise pays for itself.
BLS MI manufacturing wage distribution: P25=$19.50, median=$26.40, P75=$35.80. Company at $22 sits in the bottom quartile. The $4.40 gap to median × 273 workers × 2,080 hrs = $2.5M/yr. Current turnover at 28% × $45K/separation = $3.4M/yr in churn costs. The data suggests the externality (low wages) costs more to maintain than to internalize.
FRED shows labor force participation at 62.5% nationally — meaning 37.5% of working-age adults are not in the labor force. Our unpredictable scheduling contributes to this: drivers who leave us often exit the workforce entirely rather than take another trucking job. We're not just losing drivers to competitors — we're pushing people out of the labor market.
Assumption: When drivers quit, they go to competitors. Our scheduling practices don't affect the broader labor supply.
Track post-departure outcomes for the next 50 drivers who leave voluntarily. Survey at 30 and 90 days: Are they driving for a competitor? In a different industry? Or out of the workforce? Cross-reference against the 62.5% labor force participation rate. If our departure-to-exit ratio exceeds the national average, we're contributing to labor force dropout.
Quantify the % of ex-drivers who exit the labor force entirely. If >25% of departures result in labor force exit (vs. ~15% national baseline), document the externality: unpredictable scheduling is a labor force participation suppressor.
FRED labor force participation: 62.5%. OH transportation sector flat (3% momentum). Company pays 72nd percentile ($7.90 above median) but has 35% turnover. The data creates a paradox: strong wages + high turnover = non-compensation externality. If departing drivers exit the labor force rather than switching employers, the company is contributing to the 37.5% non-participation rate — a macro-level externality driven by micro-level scheduling decisions.
FRED data shows interest rates at 4.33% while TN healthcare employment grew by 16,800 jobs — the strongest sector momentum in our cohort (6%). We're using the interest rate as justification to NOT expand, while the market is telling us demand is accelerating. The community externality: 40,000 residents without primary care access while we sit on capacity to serve them.
Assumption: The interest rate environment makes any capital expenditure irrational. We should wait for rates to drop.
Build a scenario model using actual FRED data: interest rate at 4.33%, healthcare employment growing 6%, TN unemployment at 3.8% (strong consumer base). Model three expansion structures: (1) traditional build at 4.33%, (2) lease-to-own, (3) employer-partnership pre-commitment. Include the community externality as a quantified input: 40,000 underserved × estimated preventable ED visits × average ED cost.
At least one model shows positive NPV within 36 months. If employer partnerships can guarantee 40%+ of patient volume pre-opening, the interest rate premium is absorbed by demand certainty. Present to board with FRED data as evidence that waiting for rate drops ignores accelerating demand.
FRED: interest rate 4.33%, TN healthcare employment +16,800, sector momentum 6%. The data reveals a tension: macro conditions (rate pressure 41%) argue for caution, but sector momentum argues for action. At 3.8% unemployment and strong labor participation, the patient base is stable and growing. The externality (40K underserved residents) compounds every month of delay — emergency department costs, preventable chronic disease progression, lost productivity in the community.
FRED reports real wage growth at 1.2% nationally, but our internal data shows a bifurcation: RN wages grew 4.2% (above $42/hr, 73rd pctl) while MA/tech/admin wages grew 0.3% ($16-22/hr). BLS TN healthcare P25=$22, meaning our lowest-paid workers sit at the floor of the market. The real wage growth statistic masks the externality — our growth accrues to the top while the bottom stagnates.
Assumption: We've given raises across the board. Wages are keeping up with inflation for all roles.
Disaggregate internal wage growth by role tier against BLS P25/median/P75 benchmarks. Map each role to the wage distribution. For roles below P25 ($22/hr), calculate the gap between their real wage growth (0.3%) and inflation. Quantify how many workers are losing purchasing power while the company reports 'average' wage growth.
Produce a role-by-role externality map showing which positions are falling behind the market. If >30% of employees are experiencing real wage decline despite company-wide 'growth,' the aggregate statistic is masking an internality failure. Present the bifurcation to leadership with a targeted remediation proposal for sub-P25 roles.
FRED real wage growth: 1.2% nationally. BLS TN healthcare: P25=$22, median=$33.50, P75=$48. Company average $42/hr (73rd pctl) masks internal bifurcation. The aggregate metric hides the externality: RN wages track above median while support staff wages cluster at P25. Workers at the bottom of the distribution experience declining real wages — an externalized cost of the company's compensation structure that shows up as healthcare access inequity, turnover in support roles, and community economic drag.
Our wage structure contributes to the economic fragility of the Grand Rapids community. Employees earning below the benefits cliff spend less locally, reducing the multiplier effect that sustains the businesses around us.
Assumption: Community economic impact of our wage decisions is not measurable or material to our business.
Partner with Grand Valley State University to measure the local economic multiplier of our wage increase experiment. Track employee spending patterns pre/post wage increase.
Demonstrate that each $1 increase in floor wages generates $1.40+ in local economic activity. Build the business case for community-as-stakeholder.
We cannot fill skilled machinist positions. Open roles stay unfilled for 90+ days.
Assumption: The local labor pool lacks qualified candidates.
Raise starting wage by 12% for machinist roles. Measure time-to-fill over 60 days.
Time-to-fill drops below 45 days. Application volume increases 30%.
Training investment is wasted — new hires leave within 6 months.
Assumption: Competitors are poaching our trained workers with higher pay.
Implement 12-month retention bonus structure with quarterly payouts.
6-month retention rate improves from 65% to 80%.
Driver turnover is 35% annually despite paying above market.
Assumption: Drivers leave for higher pay at competitors.
Offer flex scheduling (4x10 vs 5x8). Measure turnover over 90 days in pilot region.
Pilot region turnover drops below 25%. Driver satisfaction improves on survey.
Route inefficiency is increasing fuel costs by ~15% YoY.
Assumption: Current routing software cannot optimize for the new delivery density.
Pilot AI route optimization on 20% of fleet for 30 days.
Fuel cost per mile drops 8% in pilot group.
Cannot expand to second clinic location due to financing costs.
Assumption: Current interest rate environment makes expansion NPV-negative.
Model expansion with phased buildout (lease-first) vs. full build. Compare NPV at current and projected rates.
Identify a financing structure where expansion is NPV-positive within 18 months.
We are losing RNs to hospital systems that offer better benefits.
Assumption: We cannot match hospital benefit packages as a small employer.
Offer tuition reimbursement program for NPs. Measure RN retention and NP pipeline over 6 months.
RN 12-month retention improves from 72% to 85%. At least 3 RNs enroll in NP program.