Overview

This final synthesis asks how government buyers can broaden access without treating small size as proof of financial constraint or productivity. In a calibrated expenditure-neutral experiment, raising the share of suppliers with 20 or fewer employees from 67% to 77% raises private output 0.41% but lowers public output 9.55% and GDP 0.79%. Contract splitting, youth targeting, and better productivity screening reduce modeled trade-offs in different ways; these are long-run model results, not universal policy forecasts.

Learning outcomes

In the calibrated size-targeting experiment, moving small suppliers from 67% to 77% raises private output 0.41% but lowers modeled public output 9.55%.
Contract splitting and youth targeting reduce modeled GDP losses relative to simple size targeting, but they change different supplier and capacity margins.
Effective access policy targets financial constraint while screening productivity and measuring the public output citizens receive.

Keywords

ProcurementSuppliersAccessProductivityScreeningOutput

Buying Small Without Buying Blind

Public procurement is the subject, and the title's question can now be answered: a government customer acts like financing when a contract creates credible future earnings that lenders can recognize, allowing a firm to borrow or accumulate wealth beyond what its existing assets permit. This final synthesis focuses on the harder judgment that follows—how a buyer should broaden access without confusing financial constraint with productive quality or ignoring the public output being purchased. The evidence supports a conditional design principle, not a universal instruction to buy from either the largest or the smallest firms.

The answer to the title is conditional

A procurement contract has two identities at once. It is a demand commitment from the government, and it can be a financial asset in the practical sense that expected public earnings affect borrowing capacity. The Spanish evidence shows the first measurable consequence: after firms obtain procurement, total credit rises persistently, with the immediate response concentrated in non-collateralized credit and collateralized credit appearing later. The model interprets that sequence through earnings-based and asset-based borrowing constraints. Public earnings are inferred to be more pledgeable than private earnings, so a government award can help a firm finance production before it has accumulated new equipment or other tangible collateral.

But financing is not the same as free capacity. The winning firm still has to deliver the order. In the Spanish data, total sales rise by around 9 percentage points, while private-sector sales fall at impact and later recover above their earlier level. The contract can therefore relax a lender's concern while tightening the firm's operating allocation: workers, machines, and working capital must be directed toward public production. Procurement profits can later increase net worth and support private-sector expansion, but the return is earned through a risky operating path rather than delivered as a guaranteed loan.

The Census evidence adds the missing access perspective. When states removed affirmative-action preferences and related support from contracting, MWBE contractors gradually became smaller relative to non-MWBEs, and fewer new MWBE contractors entered. Existing MWBEs were not detectably more likely to exit. That combination shows that government access can be embedded in a firm's scale, future opportunity, and value even when the data do not record whether a bank accepted a public contract as collateral. The case does not prove a credit mechanism, but it shows why access is upstream of any financing benefit.

The two studies should be connected, not collapsed. One follows firms after gaining procurement access in Spain; the other follows U.S. state contractors after a policy removes a particular form of access. The New York Fed study's credit estimates come from a main sample of about 14,000 Spanish firms that received at least one procurement project and from an empirical design that is not fully quasi-experimental. The Census study follows approximately 114,000 state contractors from linked restricted Census records, including about 7,000 in states that later enacted bans, and estimates relative changes around five state policy events. Their geographies, periods, outcomes, and identification strategies differ.

What they establish together is a way to reason about procurement. A public buyer changes more than the destination of a government payment. The buyer changes which firms receive a demand shock, which firms can build experience, which firms may obtain more credible earnings, and which firms can grow or enter later. The financial channel is strongest when a firm is capable of delivering but limited by credit or collateral. It is weaker or irrelevant when the firm already has ample financing. It can also be offset by delivery risk, poor productivity, or the immediate need to divert resources from private customers.

That conditionality matters for the meaning of inclusion. A smaller firm is not automatically a financially constrained firm. Some small firms are productive but asset-poor; others are small because their productivity or demand is low. A rule that treats size as a perfect proxy will reach both.

Published by DeepMox

FAQ

What did the model's buy-small experiment actually change?

It held total procurement expenditure and supplier count fixed while raising the share of suppliers with 20 or fewer employees from 67% to 77%. It was a calibrated counterfactual, not a forecast for a specific set-aside program.

Why can helping small suppliers reduce public output?

Size targeting can select suppliers with lower average productivity and net worth; with spending fixed, lower productivity can mean higher average prices and less modeled public output, even as constrained firms gain wealth.

Which procurement reforms perform better than simple size targeting?

In the calibrated model, targeting younger firms or splitting large contracts produced smaller GDP losses than simple size targeting, while productivity screening preserved more public-sector quality. These are model comparisons, not universal rankings.

Previous10 minGovernment Procurement: Shrinking Without Closing

All chaptersBack to Topic

Government Procurement: Small Suppliers and Public Output

Buying Small Without Buying Blind

The answer to the title is conditional

FAQ