US GDP Turbocharged After Record Saving Rate Revision: Why That Could Be A Problem

Last week, the US Bureau Economic Analysis unleashed some statistical goalseek magic that would make Beijing thoroughly fabricated economic "data" blush for days: as part of the comprehensive GDP revision carried out every five years by the Commerce Department’s BEA, US economic history was rewritten and in a shocking development, the annual saving rate was revised higher by 1.6% on average since 2010, with the 2017 US savings rate doubling overnight, from 3.4% to 6.7%.

Putting these revisions into context, the last comprehensive revision conducted in July 2015 only produced an upward revision of 0.3% per year. In fact, this revision to the saving rate over the prior 6 months is the largest on record, according to real-time data available from the St. Louis Fed since 1997.

Major revisions to national income (and consequently the saving rate) came from 3 sources of income: proprietors', dividend and interest income, as well as a tiny increase in employee compensation.

As BofA notes, the BEA incorporated the tabulations of sole proprietorship and partnership tax returns for 2016 which included new research by the IRS showing significant underreporting of income by nonfarm proprietors over time, causing a sizeable upward revision for proprietors' income starting in 2010. It wasn't immediately clear if this also meant a sizeable upward revision in tax audits by the IRS.

In addition, newly available IRS income data boosted dividend income in 2016 and 2017 by roughly $113bn and $143bn, respectively, shifting savings from corporations to households while the entire interest income series was revised higher due to changes in the way BEA calculates state and local employers' contribution to pension plans. Last, the saving rate got an additional boost in 2017 due to the integration of revised employment and wage data from QCEW which added $98bn to wages and salaries.

Unfortunately none of these statistical revisions actually mean anything, because both before and after the BEA revision, some 50% of Americans still don't have a single dollar in savings, and the only ones affected are those in the top percentiles of US society which means that - you guessed it - the rich get richer again.

And yet, at the statistical level, which is the only one that matters for the Fed, the latest numbers suggest households have been thrifty with their money during the recovery, and with one flick of an excel switch, the secular decline in the US personal savings rate was halted.

There are two implications from this:

  • Moving forward the higher saving rate should be a boon for the consumer BofA's economists wrote on Friday: in good times it should provide room for the consumer to spend without breaking the bank or levering up and in bad times, it should provide a buffer for the consumer against a negative income or employment shock, helping to extend the business cycle. Even so, BofA did not revise its consumer spending forecast higher significantly post revisions as the bank understands the money was never actually saved, but merely served as a plug in some big picture equation.
  • The other implication is that if the Fed was concerned about hitting an economic ceiling sometime in 2019, as a result of the roughly $500BN in incremental savings "discovered", the rate hike cycle could be far more aggressive than the market thinks.

And while BofA kept its outlook unchanged, one bank did revise its forecast as a result of the dramatic paper increase in personal savings.

In a Saturday note, Goldman chief economist Jan Hatzius quantified the implications of the higher saving rate for the consumer outlook and wrote that "taken at face value, our model implies that a declining saving rate could deliver a boost of as much as 0.8% to annualized real PCE growth over the next two years." Even under more conservative assumptions—which Hatzius writes may be warranted given the uncertainty around saving data, Goldman "now expects real consumption growth to slow more gradually than before, from 2.7% over the past year to 2.4% over the next year. "

What does this mean for GDP? Simple: it is revised sharply higher, with Goldman throwing in the towel on its warning of a sharp slowdown in economic growth in 2019 and 2020, and instead now expecting substantially higher growth as we continue drifting ever deeper into the second longest economist cycle in history.

As a result of the consumption and federal spending upgrades, we nudge up our GDP growth forecast by 0.25pp for 2018H2-2019 and lower our unemployment rate path. We still expect 3.3% growth in Q3, but now look for 3.0% in Q4 (vs. 2.5% previously), and 2.0% in 2019 on a Q4/Q4 basis (vs. 1.75%).

We now expect the unemployment rate to decline to 3.5% by end-2018, and to bottom at 3.0% in 2020. We maintain our end-2019 forecast for core PCE inflation of 2.3%.

But while Goldman's forecasting track record is just as abysmal as, well, any other economist's, the key implication is what this savings revision means for the Fed's rate hike timetable: here's Goldman's take.

We continue to expect the Fed to hike once-per-quarter until the policy rate reaches 3¼-3½% by end-2019. While the Fed is very focused on containing economic overheating, our standing forecast of quarterly rate hikes already incorporates a significant labor market overshoot. This quarterly pace seems to be largely “locked in” barring a bigger acceleration in core PCE inflation beyond 2.5%, which is a risk but not our baseline forecast. 

Our higher growth and lower unemployment projections therefore moderately reinforce the upside risks to our Fed call.

In other words, the risk to more, and faster, rate hikes is now on the table as a result of a revision in the simple calculation of how much money Americans managed to save, when in reality the vast majority of Americans have just gotten poorer no matter what the BEA's excel model says, even as "the 10%" have never been richer.

That said, it will be up to Jay Powell to decide if for the first time in 10 years, he will do the right thing, and let the market drop as a result of the Fed's tightening cycle, wiping out a substantial portion of the 10%'s net worth, when the next recession hits, or if the Fed will immediately proceed with QE4.

For now, however, the only chart worth watching is this: at what interest rate will the Fed finally cause the market to crash.