Winner of the New Statesman SPERI Prize in Political Economy 2016

Thursday 22 January 2015

That £170 billion bombshell

Paul Johnson of the IFS has written that under Labour “national debt [could be] around £170 billion higher (in today’s terms) by the end of the 2020s than would be achieved through a balanced budget.” That was all that certain newspapers needed to start talking about a borrowing bombshell under Labour.

£170 billion is a meaningless number, and the end of the 2020s is a meaningless date. First, we should put everything as shares of GDP. £170 billion is about 10% of GDP, and debt is currently around 80% of GDP. However it would be completely wrong to infer that under Labour debt to GDP would be 90% of GDP by 2030. If they achieved current balance by financial year 2017/18, then my excel spreadsheet says that with nominal GDP growth of 4% a year, by 2030 debt to GDP would be around 65% of GDP. (A few points below 65% if investment remained at 1.5% of GDP, a few points above it became 2% of GDP.) If the Conservatives balanced the overall deficit each year debt to GDP would be about 47% of GDP by 2030.

So a £170 billion bombshell actually means debt to GDP would have been reduced from 80% of GDP to around 65% of GDP. So the correct headline should have been “debt to GDP cut by a fifth in 2030 under Labour’s plans”. That is debt, which is much more difficult to reduce than the deficit. To say this is a ‘different interpretation’ is too polite – newspaper reports got it completely wrong. Who should you blame for this: Paul Johnson, innumerate journalists, biased newspapers? I’ll leave that to you.

There remains a real question of how quickly debt to GDP should be reduced. In terms of the analysis I did here and here, Labour’s plans - if it did achieve current balance by 2017/18 - are tougher than the path I described as ‘fast’ debt reduction, although not nearly as tough as Osborne’s plans. (This analysis was done before the Autumn Statement, but to pretend that the analysis needs to be revised on that account gives these numbers spurious precision.) However my ‘fast’ path did not keep to current balance after 2020, but had some further deficit reduction over the next five years. (As a result, debt to GDP was below 60% by 2030 under this fast path.) I have not seen Labour commit to sticking to current balance until the end of the 2020s. So in that sense as well the £170 billion number is meaningless.

What you should conclude from this is simple. First, as Paul Johnson and many others have pointed out, both Labour and Conservatives are aiming for tight fiscal policies (tighter than I and others think sensible given the macroeconomic situation), but the Conservatives’ plans involve substantially more cuts than Labour. Both involve reducing debt to GDP quite rapidly, so there is no question that both plans would not trouble the markets. So the only reason for going for Osborne’s plan, now apparently involving budget surpluses, is if you expect another financial crisis in the 2030s, and want debt to GDP to be something like it was before the last one. [1] Or, as a headline writer might put it (but somehow I doubt many would): “budget surpluses and austerity so we can afford to bail out the banks again soon”.

[1] For those who are really into fiscal rules, there is a technical question about whether it is better to have a target for the overall deficit or the current balance. As George Osborne has moved from the latter to the former, it may be best to read his detailed analysis of the issue. Cannot find anything? Well maybe, as I note here, he is simply following the discussion in Portes and Wren-Lewis (2014), which argues for deficit targets but a separate target for the public investment to GDP ratio.


  1. why assume 4% growth? & don't you have a problem with debt interest as the 2nd biggest govt department and growing (and ringfenced) perhaps you could write an article on that?

    1. Because the inflation target is 2%, and trend real growth before 2008 was at least 2%. I don't get the debt interest point. I have written about the benefits of reducing long run debt, but also why economic theory tells you to do it slowly.

    2. Yes: let's ignore deep recessions and take trend growth to be when the economy expands. This way we can make rosy projections and wish away debt. So easy!

    3. Assuming the present average of 3.1% servicing cost on public debt continues (which is likely an overestimate) a difference of 170bn of debt amounts to paying a difference of £5.3 bn in interest per year.

      Under the Conservative plans, then, interest payments would fall to 1.5% of GDP, while under Labour plans interest payments would fall to 2.0%, from 2.8% today.

      I am not sure why you think debt interest is the 2nd biggest government department. It is healthcare, and it will remain healthcare for quite the forseeable future.

    4. Anon, he's not ignoring deep recessions, he's using the historical record of *every* recession in every modern country to date, in which every recession has ended with a return to trend growth. Look at

      and say how *you* would sensibly project future GDP, or explain why this time is different.

    5. sorry third - i have an issue with spending more on interest than education/ defence/ all other govt departments

    6. It's not third either. The UK is also spending almost double the amount it does on debt interest, on education. It is spending more on interest than on defence, but not greatly more.

      Why, honestly, does the placement even matter, though? If we got involved a huge big pointless war, and that bumped up defence spending to higher than interest spending... Would you be pleased with that? The whole comparison game is just meaningless rhetoric. One can only note that UK spending on interest is historically actually quite low, and expected to decline regardless of whether Labour or Conservative plans are enacted, albeit not to levels as low as it was under the latter half of the (allegedly profligate) Blair/Brown government, which were the lowest they were for over 50 years..

    7. Funny how people complain that the government is spending too much on interest, and then they complain that the government isn't spending enough on interest (savers aren't getting enough interest on their government bonds because of low interest rates).

    8. @arif - the interest on my mortgage is my largest single monthly expense, but because I pay it, I get to live in a nice house. It would be even better if I was allowed to print the currency I paid my mortgage in.

  2. “budget surpluses and austerity so we can afford to bail out the banks again soon"

    The problem is we don't know where the next crash is coming from (or when it will arrive). We can probably guess that it won't be the same as last time, but that is it.

    So in your original post with deficit reduction mapped to 2079 (!) you accept, rightly I think, that it makes sense to bring the debt down, and seem to be aiming at a level of around 25% of GDP.

    An obvious objection to this sunny scenario is that it looks a bit unlikely that there won't be any shocks between now and 2079 requiring a further fiscal stimulus. Heightened debt levels leave little headroom.

    So your revised post envisages shocks every 40 years

    However, as you yourself admit, "this is complete guesswork". Indeed, Looking at the history of the UK over the last 80 years, it looks to me to be more than a bit unrealistic (especially if you think, as I do, that the lovely growth rates of the 50s are not returning to get us off the hook anytme soon).

    So, unfortunately, it looks to me like it becomes a matter of judgment incapable of mathematical assessment (ie reasonable people can disagree). How worried are we about the day after tomorrow? How much headroom do we need?

    As it happens, the short term world prospects look to me to be so grim that we will have to indulge oursleves in a bit more fiscal jam today than might otherwise be prudent.

    1. It is also interesting to note that in your world operating as it should, ie after 2095, debt only goes up as high on the peaks as around 42%.

      With our ageing population, and the natural bias of democratic governments towards running deficits, I am not sanguine about the risks.

    2. What if the time for 'further fiscal stimulus' is not in fact tomorrow, but actually today and yesterday? Isn't it silly to wait for a further crisis when we are still in one, since all the indicators suggest we are at the ZLB?

    3. I wasn't disputing that. We are running, and will continue to run for many years yet, a comparatively large (in both historic and international terms) deficit. No doubt appropriate at the zlb. The proportion of debt to GDP is rising and will continue to do so. We are getting a fiscal stimulus.

      My issue, which is more a matter of judgment, is how large the deficit should be this far after the initial crisis hit. I think S W-L's answer would be 'as large as is required to lift us off the zlb'. Looking at current debt levels, what they ought to be, the outlook elsewhere in the world, the longterm growth prospects of the UK, and the possibility of yet further shocks that may not be prudent.

      Unless you take the view that the level of debt doesn't matter (which S W-L doesn't).

    4. There is a very pragmatic reason why running the kind of policy proposed by Labour and especially the Conservatives is unwise. My paths for debt to GDP assume 4% nominal growth from 2015 onwards. That effectively assumes that by 2015 the output gap would have been largely closed, and that compared to previous trends the UK has permanently lost about 15% of real GDP. However the experience we and other countries are having with inflation suggest that this may be grossly underestimating potential. If it is, the quickest way to get debt to GDP down is to grow GDP.

      In contrast, fiscal pessimism and excessive caution are in danger of becoming self-fulfilling. We have renewed austerity, interest rates remain at their ZLB, growth remains weak.

    5. I’m baffled by Simon’s claim that “the quickest way to get debt to GDP down is to grow GDP”. Actually there’s a much quicker way, which we’ve implemented big time over the last three years or so: QE. That is, print money and buy back debt (or desist from rolling over debt, print money and pay off creditors).

      If that’s too inflationary, no problem: just raise taxes and unprint the money collected. As long as the deflationary effect of the latter equals the above inflationary effect, GDP stays constant.

      Spinning Hugo makes a similar point to Simon when he says “…the short term world prospects look to me to be so grim that we will have to indulge oursleves in a bit more fiscal jam today than might otherwise be prudent.”

      Let’s think about that. If growth is too low we do some “fiscal jam”, i.e. we “print and spend” or “borrow and spend”. Either will do (as Keynes pointed out). If there are plenty of willing purchasers of debt, then borrow and spend won’t raise interest on the debt. If there aren’t, we go for print and spend.

      Where’s the problem? Where’s the “lack of prudence”?

    6. “budget surpluses and austerity so we can afford to bail out the banks again soon”.
      Allow me a paraphrase, at least in the US situation:
      “budget surpluses and austerity so we can afford to again bail out the losses of the 1% on the next financial crisis expected soon since whatever weak reform was set up after the crisis is now being weakened further”.

  3. Ralph's point expanded: The choices are three: print, or borrow, or tax to create the cash needed to cover wise spending policy. Assuming most agree that government should be supporting aggregate demand after it trended its support the opposite way for a bit, and assuming you also believe in the reasons why progressivity is used in tax regimes (to try to balance tax burdens as a matter of fairness based on propensity to spent and utility) - then we have defined the target for nominal spending for the period in question. Keynes more or less says that you should finance with taxation when economies are stable but borrow or print more (and bury for treasure hunts,according to Keynes) when economies are in depressive periods. It is a simple matter to print currency, ex nihilo (like a bank), to finance spending and this should be done until markets signal the need to change.

    Here is the rubric representative democracies should follow: target nominal spending based on wise economics, then determine what type of economy you have now, this determination determines the mix of taxation and/or printing to finance the outlays, avoid borrowing from the wealthy unless you have a severe recession/depression.

    Tax-cut-and-borrowing from the same group of already wealthy people is a wealth-transfer-scheme (a political economy result) - that is all it is -- can't understand why this approach is used at the current time in large, relatively stable societies that have their own currency; namely, the UK, Europe, US and Japan.

    Inequality of net worth happens, tidal forces are real and impersonal too. But we do not need to use our own govts to effect wealth transfer schemes (maybe in some historic period we should do this, though I still can't imagine when or why this makes sense except as an expression of political power, can't see the economics or political science of it though).

    SWL please note that defence of the existence public debt, per se, is understandable but you have to take care with the fundamentals of the message you are presenting. I recommend you always say in your commentary that in certain times we should borrow but we should try to avoid borrowing especially if we are simply transferring wealth and ownership of future economic productivity to those who are already wealthy in order to finance support for current aggregate demand (it would be far-preferable to just print the currency) - balance the message. More people will trust it more, in my opinion.

    1. Anon,

      I agree that “Tax-cut-and-borrowing from the same group of already wealthy people is a wealth-transfer-scheme…”. That is, the larger is government debt, the higher will interest rates be, all else equal. And high interest rates obviously suit the cash rich.

      Like you, I’m skeptical as to what the merits of borrowing are as compared to printing. If borrowing was easier to reverse than printing, or if lags were shorter in the case of borrow, those would be arguments for borrowing. But as far as I can see, neither the reverse nor the lag arguments are very convincing.

    2. RM - Just to be sure, market reactions on interest rates are not the direct wealth transfer mechanism. If I am wealthy and had a tax bill of $10K and you told me I didn't have to pay this anymore but that you still wanted the cash from me and would GIVE me a tradeable security/asset instead of the tax bill the math is very simple - it is on its face a $20K positive difference in my net worth (before it was -$10K, now +$10K - a positive $20K difference for me). I love this financing scheme, give me more deficit spending this way!! And don't forget I am gaining increasing direct control over the future too, as I own more of the future income stream (whether there is growth or decline in the future, I own the income as you just gave it to me).

      Can people do this simple math????

      Are people so hung up in the angels-on-the-pin view of the CB world of accounting ledger transfers that they can't see this?

      (Side note - yes, I also agree that those who make the financial-asset trading markets and those who trade in them like interest rates to be volatile and upward, as this produces a dynamic marketplace of opportunity for them. Market-makers especially like this since they create and price credit, and can use the volatility to ratchet rates upward (this is real 'inflation' unhinged from the natural rates of real-bills). So yes cash rich like increasing interest rate periods too - but the wealth transfer scheme is a much simpler math.


Unfortunately because of spam with embedded links (which then flag up warnings about the whole site on some browsers), I have to personally moderate all comments. As a result, your comment may not appear for some time. In addition, I cannot publish comments with links to websites because it takes too much time to check whether these sites are legitimate.