Duncan’s Economic Blog

Fiscal Multipliers

Posted in Uncategorized by duncanseconomicblog on December 13, 2010

The table below is from the neutral Congressional Budget Office.

Now, as I’ve said before, multipliers are probabaly a bit lower in the UK than in the US (we have a more open economy and so spending is more likely to “leak” abroad), but this is still important.

The concept of the multiplier is simple – it measures the effects of changes in fiscal policy on the economy as a whole. A multiplier of 1.0 means that a one pound change in policy effects GDP by one pound. A mutliplier of 1.5 means a one pound increase in policy increases GDP by £1.50p.  0.5 means a £1 change in policy moves GDP by 50p.

Notice that tax cuts on low and middle earners will be more stimulating to the economy than tax cuts on high earners. Notice too that generally spending is a more effective means of getting growth than cutting taxes and remember that these multipliers work in reverse to – so cutting spending is likely to hit growth harder than raising taxes.

16 Responses

Subscribe to comments with RSS.

  1. Neil Wilson said, on December 13, 2010 at 11:35 am

    Looks like the best ‘bang per buck’ is infrastructure projects and general job creation then.

  2. Tim Worstall said, on December 13, 2010 at 12:14 pm

    Oooh, oooh, I see what you’ve done there!

    You’ve compared “temporary” tax cuts with spending.

    And we all know that temporary tax cuts have a very different effect from permanent ones…….

    • duncanseconomicblog said, on December 13, 2010 at 12:17 pm

      Here’s some OBR estimates for the UK:

      Table C8: Estimates of fiscal multipliers

      Impact multipliers

      Change in VAT rate 0.35

      Changes in the personal tax allowance and National Insurance Contributions (NICs) 0.3

      AME welfare measures 0.6

      Implied Resource Departmental Expenditure Limits (RDEL) 0.6

      Implied Capital Departmental Expenditure Limits (CDEL) 1.0

      Same picture, and not talking about temporary measures.

      • Neil Wilson said, on December 13, 2010 at 4:05 pm

        Which direction are those in? I can’t imagine anybody is expecting the forthcoming rise in VAT to do anything other than reduce GDP.

        • duncanseconomicblog said, on December 13, 2010 at 4:11 pm

          Both ways. OBR reckons £1 raised through VAT reduces GDP by 35p. And the other way, OBR reckons a £1 cut in VAT revenues increases GDP by 35p.

          (What’s nice about this is that the OBR reckons a rise in VAT has a stronger negative effect than a rise in National Insurance Contributions, the Tories choose to drop the NICS rise on employers and instead increase VAT).

          I think the OBR estimates are on the low side, although I accept that our multipliers will be lower than those of the US.

          • Neil Wilson said, on December 13, 2010 at 7:32 pm

            Raising any sort of Tax is mildly mad given the lack of demand in the economy. NICs is a lousy tax that should just be scrapped and raising VAT at the moment is plain bonkers.

            If our multipliers are that low, that’s a heck of a leakage to savings. It’s going to take a mighty push to get us out of the doldrums.

  3. Tim Worstall said, on December 13, 2010 at 12:17 pm

    Another possible response:

    “Keynesian economic model shows that Keynesian policies work best

    Film at 11!”

  4. vimothy said, on December 14, 2010 at 12:03 pm

    Duncan,

    Can you give us a bit of an idea about how the CBO generated these estimates?

    • duncanseconomicblog said, on December 14, 2010 at 12:09 pm

      Pages 14 to 16 of this PDF from the CBO.

      http://cbo.gov/ftpdocs/106xx/doc10682/11-30-ARRA.pdf

      • vimothy said, on December 14, 2010 at 1:25 pm

        Thanks Duncan, but I don’t feel that much more enlightened to be honest. Are they saying its a mix of econometric modelling and DSGE simulations?

        I ask because everyone seems to get such wildly different values for these–seems heavily model dependent.

        • duncanseconomicblog said, on December 14, 2010 at 1:30 pm

          As far as I;m aware all DSGE and econometric modelling.

          Some good discussion around multipliers here:

          http://www.imf.org/external/np/g20/pdf/031909a.pdf

          The estimates do vary quite alot – although they all support the point I was trying to make that you can rank them as= capital spending (highest), other spending, taxes.

          And that money given to the poorest (as welfare payments or tax cuts) has a higher multiplier than money given to the richest.

          • Tim Worstall said, on December 14, 2010 at 1:40 pm

            “The estimates do vary quite alot – although they all support the point I was trying to make that you can rank them as= capital spending (highest), other spending, taxes.

            And that money given to the poorest (as welfare payments or tax cuts) has a higher multiplier than money given to the richest.”

            This is hardly surprising, given that the models used to generate the rankings assume this as part of the construction of the model.

            A parallel would be for me to assume when constructing a model that the peak of the Laffer Curve is income tax at 40% and then expressing surprise when my model spits out that tax rates over 50% lose revenue.

            • vimothy said, on December 14, 2010 at 1:51 pm

              Tim,

              Can you talk me through what you mean, with ref to a simple model actually used by one of these agencies?

              • Tim Worstall said, on December 14, 2010 at 2:00 pm

                No🙂

                Beyond my technical capabilities.

                But it’s hardly surprising that a model constructed using Keynesian assumptions comes up with Keynesian conclusions now, is it?

  5. vimothy said, on December 14, 2010 at 2:11 pm

    Well, if your premise is correct, then yes, but…

    But waddya mean by “Keynesian assumptions”? And are the CBO’s models actually constructed using them? How do you know? Are we talking about the econometric models here or the simulations?

    Questions, questions…

    • vimothy said, on December 14, 2010 at 2:12 pm

      Sorry, luddite here trying to respond to Tim at on December 14, 2010 at 2:00 pm


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: