According to the latest weekly Factset data, after five consecutive quarters of EPS declines, earnings for the Q3 season, which begins in earnest this week, are set to post a decline in of 2.1%, which would mark the first time the index has recorded six consecutive quarters of year-over-year declines in earnings since FactSet began tracking the data in Q3 2008. It also means that the vast majority of S&P500 firms are in a buyback blackout period for the next 3-4 weeks.

And while there has been a pickup in recent optimism that Q3 earnings will actually surprise to the upside, and post a material beat relative to consensus as they traditionally have in past years, one firm that is pouring cold water on hope for a rebound is Goldman Sachs, whose David Kostin writes in his latest weekly kickstart, where he previews Q3 earnings, that "changes in most macro factors point to disappointing results." and points out that 4 out of 5 macro variables suggest that third quarter earnings will disappoint. To wit:

Negative changes in most macro variables suggest 3Q earnings will disappoint relative to expectations. The conclusion is based on our analysis that utilizes five macro factors that have historically been correlated with earnings surprises: US economic growth, interest rates, oil price, the dollar, and consensus EPS revisions. Of the five factors, four have shifted in a direction that typically weighs on earnings surprises. Oil price change is the only factor that should positively contribute to 3Q surprises. During the last 15 years (60 quarters), an average of 46% of S&P 500 firms reported EPS results more than one standard deviation above consensus estimates. Our model suggests that a modestly below-average share of firms (43%) will post positive EPS surprises in 3Q.

Here are the details why Goldman continues to be notably gloomy on corpoate prospects, a trend we noticed first in the late spring, and which has gotten worse since then:

3Q 2016 earnings preview

Earnings season officially begins next week when Alcoa (AA) reports, but a handful of S&P 500 firms have already announced results. Notable earnings beats include COST, PEP, and FDX. Of the small share of companies that have reported 3Q earnings so far, 64% have announced EPS more than one standard deviation above consensus estimates, compared with the historical average of 46%.

The vast majority of S&P 500 firms are now in the buyback blackout window. Roughly 66% of S&P 500 market cap will announce results during the next three weeks, and 31% of market cap will report during the three day span from October 25-27. Nearly 90% of S&P 500 market cap will have reported earnings by November 4 (see Exhibit 1).

Consensus bottom-up forecasts imply 3Q S&P 500 adjusted EPS will fall by 1% on a year/year basis. EPS declines will be driven mostly by the Energy sector. Energy EPS is forecast to decline by 68% year/year, but outside of Energy, S&P 500 EPS is expected to rise by 3%. Analysts predict S&P 500 sales will grow by roughly 4% but be offset by margin declines (see Exhibit 2).

Aggregate S&P 500 margins are projected to fall by 59 bp in 3Q. Energy sales should decline only modestly, meaning the massive sector-level EPS decline can be attributed primarily to margin compression (-384 bp). Outside of Energy, consensus forecasts imply that margins will fall across all but two sectors and that S&P 500 ex-Energy margins will decline by 27 bp. Only Materials sector margins are expected to rise, and Information Technology margins are currently expected to remain flat.

Negative changes in most macro variables suggest 3Q earnings will disappoint relative to expectations (see Exhibit 3). The conclusion is based on our analysis that utilizes five macro factors that have historically been correlated with earnings surprises: US economic growth, interest rates, oil price, the dollar, and consensus EPS revisions. Of the five factors, four have shifted in a direction that typically weighs on earnings surprises. Oil price change is the only factor that should positively contribute to 3Q surprises. During the last 15 years (60 quarters), an average of 46% of S&P 500 firms reported EPS results more than one standard deviation above consensus estimates. Our model suggests that a modestly below-average share of firms (43%) will post positive EPS surprises in 3Q.

Higher economic growth increases the likelihood of positive EPS surprises. However our Current Activity Indicator (CAI) suggests a 72 bp decline in the pace of US economic growth relative to the third quarter of last year, a factor that is likely to weigh on S&P 500 EPS results. The 10-year US Treasury yield averaged 1.6% in 3Q, 66 bp lower than the quarterly average one year ago. At the index-level, falling interest rates lead to fewer positive earnings surprises by hampering results in the Financials sector. Looking beyond the macro lens, our Banks analysts highlighted decelerating C&I loan growth as a key theme for 3Q earnings. Slowing C&I loan growth will pressure aggregate loan growth, which consensus still forecasts will be above the three-year average. KEY, FITB, and PNC are most at risk (see 3Q16 Preview, October 5, 2016).

Consensus EPS revisions during the 3-month period preceding earnings season are directionally indicative of earnings surprises. Consensus bottom-up estimates for 3Q EPS fell by 3% throughout the quarter, foreshadowing a below-average earnings season. Since June, consensus has lowered 3Q EPS forecasts in most sectors, with Energy EPS slashed by more than 30%. A disappointing 3Q earnings season will spark negative revisions to 4Q EPS estimates, which are overly optimistic in our view. Consensus bottom-up forecasts currently suggest 4Q EPS will grow by 7%, representing the highest year/year growth rate in nearly two years. Changes in the price of crude oil, which has fallen by 9% since 3Q 2015, is the lone macro factor that should positively impact the 3Q earnings season. Brent crude price has jumped 88% since bottoming in January, but the average price during 3Q 2016 was still down near-double digits year/year. Declining oil prices will weigh on Energy firms but should continue to serve as a modest benefit to the earnings of companies in other sectors.

Finally what does this all mean for the market:

While cyclicals have outperformed defensives by 9 percentage points during the last three months (7% vs. -2%), a disappointing earnings season is one reason why the outperformance will not persist. S&P 500 has traded between 2100 and 2200 for almost three consecutive months (64 trading days). As the stock market stagnated, bond yields started to rise. The 10-year Treasury yield has climbed by 38 bp since July, while defensives (GSSBDEFS) have lagged cyclicals (GSSBCYCL) and the S&P 500. The recent strong performance of cyclicals interrupted a nearly two-year stretch of outperformance by defensives. A poor earnings season could pressure stocks, sending the market index lower and pushing investors back toward defensives. Additional negative EPS revisions and rising uncertainty in advance of the presidential election add to the case for defensive firms.

Alternatively, a poor earnings season and a drop in stocks could force more dovish talk out of the Fed, killing the prospect of a December rate hike, resetting the cycle once again and leading to another rebound in stocks driven entirely by multiple expansion.