Last revised 2012-03-08
As usual, our methodology is to rate the bills, regardless of how the individual legislators voted on them. Most bills are rated from -2 (most hostile to liberty) to +2 (most favorable). Last year, we added -3 and +3, for particularly egregious and excellent bills, respectively, to the weighted ratings. Associated with the rating is a comment indicating why the rater rated the bill. The spreadsheet applies the bill rating to each vote to obtain a rating for each legislator. These are then summed.
The spreadsheet is generated with a custom Perl script, new in 2010 and revised yet again for this year. The script assigns votes for (1) or against (-1) a bill on third reading to each legislator. Raw data come from the digests for each bill on the Legislative Services Office (LSO) web site. In the case of the 2012 session, the raw data come from the 2012 summary.
Similarly, the spreadsheet sums sponsorships and accumulates them into the ratings.
We indicate whether a bill is signed into law or not, but do not rate the governor.
We provide an outcome code for each bill:
These are summed up on each worksheet, then the worksheet sums are accumulated in the Results worksheet.
We provide the LSO's summary (which are sometimes opaque and more technical than the casual citizen would like) in comments. Also, after some bills are amended, the LSO summary is out of date.
We make an effort to indicate fiscal impact of a bill by scraping the LSO's Fiscal Impact statements. These, too, are not updated as the bill is amended. If possible, we apply one of several indicators:
The comments in the fiscal impact row may contain more information. Occasionally, a bill's fiscal impact statement is missing from the web site when we scrape it. In that case, the comment will say, "No fiscal data available."
The budget is handled in a special way. Each house starts with a budget bill, usually SF1 and HB1. These are identical "mirror" bills. They are amended and voted in parallel in each house. They are then reconciled in committee, and the results submitted back to the two houses. Thus only the House has a third reading vote on HB1, and only the Senate on SF1. Joint Rule 14 (PDF) The spreadsheet reflects that special handling.
This year there are again 13 categories for bills, carried over from last year. They are in flux and may change in future. These are rough categories, and different reviewers may assign a bill to different categories. You can find the categories for each bill in the comments for each rating, and in the bill's summary PDF. Each category has a row in the spreadsheet below the legislator rows; there is a 1 for each bill in a category. These are summed on the Results worksheet.
The categories are:
The categories also show up on the legislator pages. We show each legislator's score in each category, as though you had created a spreadsheet with only the bills in that category. Note that the average is an average of the per category scores. It is not the same as the overall score in the spreadsheet, and one should be surprised if those two values happen to be equal.
The Liberty Index is not intended to be an exact rating. For one thing, it only looks at third reading votes, not at votes on amendments or in committee. In a budget session, a two thirds vote in the house of introduction is required to introduce a bill other than the budget. Clearly these votes affect the fate of a bill. In spite of the precision with which these numbers are displayed, they are at best rough guides. Also, categories may not map well from year to year as categories may come and go. House and Senate ratings are not comperable, as the two houses vote on different bills.
2011 provides an interesting example of the variability of rating bills. HB 41 (PDF) received from three different raters a 2, a 1 and a -1, for a net rating of 2. Take ratings of any one with your salt shaker handy.
Once again, it is interesting to see the results, as members of the legislature seem to track consistently from previous years, despite the changeover in bill raters. The conclusion seems to be that, whatever the differences in political philosophy, people apparently have a fairly consistent understanding of what liberty is.
The addition of our "Dr. No" hypothetical legislator is interesting. Some of the real legislators do better than Dr. No, but most do worse. While one year of data is hardly definitive, it suggests that we can add to our advice to legislators, "When in doubt, vote to kill."