Research & PolicyFree
Regulatory Impact and Cost Analysis Tool
Send the rule text and who it reaches, get the cost by affected group with each group's burden as a share of revenue.
River builds the cost estimate one affected group at a time and reports each group's burden against its own revenue, not just the total. Fixed costs get separated from variable ones, because a fixed compliance cost is what makes a rule regressive across firm size. Then it prices the thresholds: who the exemption excludes, what the entity immediately above the cutoff pays, and how far the cliff moves the decision. The aggregate figure is still there, and it is no longer the only figure.
Search this and you get templates. A table with columns for costs, benefits and net impact, a heading for affected stakeholders, and a worked example in round numbers. The federal guidance that governs this work asks for something a table of totals cannot hold. It says analysis should provide a separate description of distributional effects, meaning how both benefits and costs land across sub-populations, so a decision maker can weigh those alongside efficiency.
Written for the analyst drafting a comment, the association costing a proposal for its members, and the researcher who needs a defensible number rather than a plausible one. It expects the coverage question already settled, which is what reading a proposed rule from its definitions outward does, or applying a bill's amendments to the statute they edit for legislation. The result lands in a brief that carries its own evidence or a filing in the docket. Where another jurisdiction already tried this, the grid with a source in every cell is the place to check.
The average agency pays nine thousand dollars
A state proposal reaching 1,840 licensed home-care agencies. Setup is $4,200 plus $18 per field worker; recurring is $2,400 plus $9 per worker plus a $3,600 annual audit. Annualising the one-time cost over three years at seven per cent gives a factor of 0.381. Aggregate annualised cost comes to $17.0 million, or $9,259 per agency per year. That is the number a summary reports, and on its own it is close to useless.
Break it by band. Agencies under $500,000 in revenue pay $7,822 a year, which is 2.52 per cent of revenue. Agencies over $10 million pay $21,873, which is 0.078 per cent. The larger firm pays 2.8 times more in dollars and 32 times less as a share of what it earns. The $9,259 average is 2.99 per cent of a small agency's revenue and 0.033 per cent of a large one's, so the aggregate figure is wrong for everybody it describes.
The distribution of the total runs the other way, which is the part that surprises people. The smallest band bears 34.1 per cent of the aggregate cost and the largest bears 12.3 per cent, because there are 742 small agencies and 96 large ones. Small entities are 1,360 of 1,840 firms, 73.9 per cent, carrying 64.3 per cent of the cost. Then the cliff: at nine field workers an agency pays nothing, at ten it pays $7,759, which is 3.53 per cent of a $220,000 revenue.
How it works
Fix the population
Who the rule reaches, counted, and broken into the groups whose burden will differ.
Separate the costs
Fixed against variable, one-time against recurring, each carrying the source that produced it.
Compute per group
Cost per entity in each group, then that cost against the group's own revenue.
Price the thresholds
Who the exemption excludes, and what the entity just over the line now pays.
What you get
- Cost per affected group, and each group's burden as a share of its own revenue
- Fixed costs separated from variable ones, since fixed cost is what makes a rule regressive
- Which share of the aggregate each group bears, alongside what it costs them proportionally
- The exemption threshold priced, including what the entity just above the cutoff pays
- Every assumption carrying its source, and every missing input named as missing
- A small-entity count and burden, in the form a flexibility analysis has to state
Common questions
Why is the burden as a share of revenue the number that matters?
Because a compliance cost is only a burden relative to what an entity earns, and fixed costs do not scale down. In the worked example the largest firms pay more in dollars and 32 times less proportionally. A rule described by its aggregate cost looks even-handed and can still fall almost entirely on the smallest entities it covers.
How does it handle an exemption threshold?
By pricing both sides of the line. It reports how many entities the exemption excludes and what the entity immediately above the cutoff pays, which is the full fixed cost with none of the scale. That cliff is often the most consequential design choice in a rule and the least analysed, because a table of totals cannot show a discontinuity.
What if I do not have all the cost inputs?
Then the missing ones stay visible. Each is named, with what it would change, and the estimate is reported as a range with the gap stated rather than a point built on a guess. An estimate with two named holes is more useful in a comment than a complete-looking number a reviewer can take apart.
Can it do benefits as well as costs?
It states benefits where they are quantifiable and their incidence where it is not. Benefits and costs frequently land on different people, which is the whole reason the distributional description exists. Where a benefit cannot be monetised honestly the analysis says who receives it and how it would be measured, rather than assigning it a manufactured dollar value.
Does it produce the small-business analysis a filing needs?
It produces the substance. An initial regulatory flexibility analysis must contain an estimate of the number of small entities to which the proposed rule will apply and a description of the compliance requirements falling on them. The group-level output is exactly that, with the counts and the per-entity burden already separated.
The agency published its own estimate. Can I compare?
Yes, and that comparison is often the most persuasive thing in a comment. The run puts your figures next to the agency's line by line and isolates where they diverge, which is usually one assumption such as staff hours per submission. Then the dispute is about a single number with a source, rather than two totals.
What comes back?
A Sheet holding the cost model with a row per affected group, every assumption in its own labelled cell, and the burden calculation exposed so a reader can change one input. A Doc writing the analysis with the distributional effects stated, the uncertainty quantified, the thresholds priced, and the inputs you did not have named.
Regulatory Impact and Cost Analysis Tool
Fill in the form and your workspace opens with the work already underway.