The Budget Browser: A Resource for Everyone

First, I am very glad that our new tool is helping to fuel these types of debates and that it is making data available to people on all sides to examine, analyze, and make their case.  Informed debate about the tax and spending decisions that we as a state make is at the heart of real democracy.  

Before addressing the technical issues in this debate, I should note that none of these technical matters answers a more fundamental question:  whether it is better for society to increase or to decrease the level of resources we dedicate to education, public health, human services, health care, support for local services, and the other things that we do through government.  The Browser allows people to look at spending trends in each of these areas (and many others) and to draw their own conclusions about these trends.

In these two posts http://bluemassgroup.com/diary/19727/charlie-bakers-negative-attack-on-patrick and http://www.redmassgroup.com/diary/8363/fuzzy-math-over-at-bmg-on-devals-spending, posters on Red Mass Group and Blue Mass Group make seemingly contradictory claims about state spending growth (both relying, at least partially, on the Budget Browser).  So who’s right?  At the heart of the debate is a question about how to account for inflation when comparing costs over time.  Should we account for inflation, and if so, how?

The Browser offers three options for adjusting for inflation.  One is to look at data in nominal dollars with no adjustments to account for inflation.  While it is rarely appropriate to compare spending from different time periods without adjusting for inflation, we believe that it is important to provide this unadjusted data so that people who want to use some type of adjustment other than those built into the Browser can use the raw data to do so (as Rob “EaBo Clipper” Eno did).  

The second option offered by the Browser is to adjust using the consumer price index (CPI).  This is a common measure of inflation.  The Bureau of Labor Statistics (BLS) calculates the CPI by constructing a “market basket” of the goods and services a typical consumer buys and then tracking the changes in price of those goods and services.  While this is often an effective inflation measure to use when comparing costs over time, it is not ideally suited for use for comparing overall government spending over time.  The problem is that the things that governments buy are different from the things that consumers buy.

To address this problem, the Bureau of Economic Analysis (BEA) reports another measure of inflation which tracks the changes in costs of the things that state and local governments buy.

The Browser allows users to decide which measure of inflation makes the most sense and to apply that measure.  Different choices about what measure to use account for most of the differences in the posts by Rob Eno and JohnK.  There is more information about these inflation issues in a memorandum prepared by the New England Public Policy Center at the Federal Reserve Bank of Boston. Available here.

Rob “EaBo Clipper” Eno suggests that the consumer price index or various other measures should be used for FY 2008 and FY 2009 and that no inflation adjustment should be used for FY 2010 (because FY 2010 isn’t over we don’t know what inflation will be). He makes inflation-adjusted comparisons for FY 2008 and FY 2009 but not FY 2010. It is correct that we don’t know what inflation will turn out to be in FY 2010.  Our Browser uses the most authoritative projections available.  But we also provide the data in nominal dollars (not adjusted for inflation at all).  Rob Eno also adjusts using various measures of economic growth.  This can be a useful way to look at changes over time, but when doing so it is important to compare changes over a complete business cycle (in periods of strong economic growth these measures would make cost growth appear smaller and in recessions would make it appear larger).

JohnK at Blue Mass Group presents the data adjusted for inflation using the measure of inflation that tracks the costs of the things governments buy.  This is the measure that most directly adjusts for the effects of inflation on government costs.

The other difference is that Rob Eno adjusts the FY 2007 budget data to subtract out the spending restored to that budget by Governor Patrick which had been cut by Governor Romney using 9C authority. The Budget Browser shows only the final budget for past years (the budget changes numerous times each year as supplemental budgets are approved and 9C or other cuts are implemented).   Because the primary purpose of the Browser is to let people compare budget trends over time we don’t post every version of each year’s budget.  But in the context of comparing trends during two different administrations, it is not unreasonable to make the adjustment that Rob Eno suggests.

So what do we see if you adjust FY 2007 down by the amount that Governor Patrick restored from 9C cuts?  If you don’t adjust for inflation, spending increased between FY 2007 and FY 2010.  If you do adjust for inflation in the costs of things that governments buy, then spending declined between FY 2007 and FY 2010.  Spending also declined during those years using the CPI as the inflation adjustment. Spending did increase in real terms between 2007 and 2009 – but spending did not increase for the full period from 2007 to 2010, because there were significant spending cuts in 2010.

I hope this clears up some of the apparent conflicts.  I do encourage everyone to use the tool to dig deeper.  You can look not just at overall numbers, but also at specific categories and sub-categories.  The Browser has data going back to 2001 so that you can look at longer term trends and see if you think the choices that have been made in budget and tax policy (see the Revenue section) have been the right choices or if there could have been — or could be — better choices.

— Noah Berger, President, Massachusetts Budget and Policy Center

Please share widely!

0

0