# Nominal spending growth and real service inputs

Data Note · Public Finance
A rise in nominal public expenditure establishes that more money was recorded as spent. It does not determine how much additional labor, equipment or public service that money purchased. The missing information concerns the composition of spending, the prices of its inputs and the relationship between inputs and delivered services.

Insee's [discussion of nonmarket activity measurement](https://blog.insee.fr/sante-education-services-administratifs-la-difficile-mesure-des-activites-non-marchandes-en-temps-de-crise-sanitaire/) distinguishes expenditure that supports recipients' income from expenditure used to produce public services. It also explains why the absence of an economically significant selling price complicates the valuation and volume measurement of nonmarket output. Its account concerns national-accounts methods, not a claim that every increase in a French budget corresponds to more services.

## A larger wage bill can purchase unchanged labor input

Consider a hypothetical service provider initially spending 60 on employees and 40 on purchased operating inputs. Total expenditure is 100. In the following period, wage rates and the prices of those purchased inputs each rise by 10 percent, while quantities and workforce composition remain unchanged.

The provider now spends 66 on employees and 44 on other inputs, for a total of 110. Nominal expenditure grows by 10 percent, but the stated input quantities do not increase. Deflating the two components by their matched price relatives returns 60 and 40.

Using a general consumer-price increase of 2 percent instead would produce 110/1.02, or approximately 107.84 in initial-period units. Describing the difference as a 7.84 percent rise in service provision would impose a price index that does not match the hypothetical provider's costs. The problem is the deflator, not the division itself.

Actual public services buy different combinations of labor, buildings, energy, medicines and other goods. Their relevant price movements need not match one another or the consumption basket of an average household. A change in the spending mix can therefore affect the aggregate input-price calculation even when component price rates are stable.

## Input volume is not delivered output

Suppose, in a separate hypothetical case, a provider uses unchanged measured inputs but delivers 105 comparable treatments instead of 100. An activity-based output measure would rise by 5 percent if treatment definitions, quality and case composition were unchanged. An input-based measure would show no volume increase.

This difference cannot be resolved by choosing the larger number. The first requires evidence that the treatments are comparable and that the count captures the service delivered. The second records resources used but may not detect a change in efficiency. More complex cases or changes in quality can alter either interpretation.

Insee discusses distinct methods for individual and collective nonmarket services. The relevant lesson for an expenditure briefing is to establish which service-volume measure is available and how it is constructed, rather than infer a general output change from the nominal budget alone.

## Keep transfers outside the service-input calculation

A cash transfer to a household can be important for the recipient's resources without representing a directly produced public service of equal value. Adding such transfers to an operating budget and deflating the combined total as though it purchased one homogeneous service confuses different transactions.

Capital expenditure introduces another timing distinction. Buying a building or piece of equipment can support service production over several years. The purchase amount is not automatically the value of services delivered in the purchase year. Utilization, operating staff and the timing of availability all affect that connection.

For a French expenditure assessment, the useful progression is from nominal transactions to matched input prices and quantities, then to measured service output where defensible indicators exist. Each stage answers a different question. Maintaining those distinctions makes it possible to discuss fiscal effort, resource availability and public-service delivery without claiming that one aggregate has measured all three.

---

American Economic Senate