US retail sales ticked higher in August, rising 0.2%–below several consensus forecasts but still a decent if unimpressive gain. Industrial output in the US in August, on the other hand, was clearly disappointing, suffering a 0.4% slide vs. the previous month—the weakest performance in three months. On a year-over-year basis, industrial activity also weakened, with growth dipping close to its lowest pace since the US recession ended in 2009. Does this add up to a clear signal that a new recession for the US is now fate? Some are tempted to make that call, but I’m not there yet, as I’ll explain.

Macro risk is certainly elevated, as I’ve been discussing for weeks (see here, for instance). But it’s still hard to quantitatively distinguish between what may turn out to be a run of slower growth vs. the start of an NBER-defined recession. For some, that’s a false distinction, but that’s an issue for another day. Meantime, let’s review today’s numbers from the comparatively reliable noise-filtering prism of rolling one-year percentage changes, starting with retail sales.

The headline data for consumer spending last month dipped to a 2.2% increase from the year-earlier level. Here too the trend is close to its weakest growth rate in six years. That’s a worrisome sign, but hardly fatal at this stage. Why? One reason is that when we exclude gasoline sales, the annual gain for retail spending is substantially higher, rising 4.4% a year through August. That’s below the pace that prevailed for much of last year, but it’s still a moderately healthy rate. The main takeaway: consumption excluding gasoline sales, which are lower due primarily to falling energy prices, remains relatively stable and rising at a moderate trend rate. Note the upward sloping trend line in the chart below for the trailing 24-month period.

Industrial activity, by contrast, looks noticeably weaker in annual growth terms. Output is still advancing, but at a sluggish pace: higher by just 0.9% for the year through August. That’s effectively a match with June’s 0.8% year-over-year increase—the lowest rate in six years. The manufacturing component’s a bit stronger, but here too the trend has turned unusually soft.

The discouraging numbers for industrial activity aren’t much of a surprise at this point. It’s been clear that manufacturing growth has decelerated this year, as Markit’s PMI flash release for this month points out. Today’s disappointing September data for this sector in the New York Fed region is another clue for thinking that the weakness will roll on.

The question is whether the weak manufacturing sector and softer but otherwise stable consumer spending is enough to trigger a new US recession? By some accounts, the answer is an emphatic “yes.” That’s a tempting conclusion if you extrapolate the recent downward sloping trend for some indicators—industrial output in particular–into the months ahead. But there are still data-based reasons for reserving judgment on what comes next.

Exhibit A is the labor market, which is still expanding at a decent year-over-year rate—private non-farm payrolls increased 2.4% in August vs. the year-earlier level. Here too the trend is softer vs. recent history, but the encouraging numbers in jobless claims lately offer support for arguing that employment growth isn’t about to fall off a cliff.

There’s also the Fed factor to consider. Given the latest round of numbers, it seems likely that the central bank won’t announce a rate hike on Thursday. A minor point in the grand scheme for macro, perhaps, but a delay in squeezing policy would at least reduce concerns that a monetary error is upon us.

The bottom line: the macro trend for the US has obviously weakened. It’s unclear if the deceleration will persist. Perhaps the key question now: Will the troubles in manufacturing remain contained–or spill over into the wider economy? If recession is on its way, the critical factor will be an expansion in the deterioration into the realm of payrolls. But that’s not happening in a material degree based on the available data to date. October and beyond, of course, is up for debate.

Overall, the risk that the US business cycle is slipping off the edge is higher, but only moderately so and still short of the point of no return. Nonetheless, we’re in a precarious period when a run of disappointments could tip the scale rather quickly. We’re not there yet, nor is it clear that we will be in the near term–I’ll have more to say on this when I update the Economic Trend & Momentum indexes for September on Friday (Sep. 18).

For now, there’s a modestly higher possibility that trouble is lurking. The Capital Spectator’s macro projections still imply that we’ll dodge this bullet. But the margin of comfort has narrowed. Modest growth prevails so far, and that’s still a reasonable forecast… for now. But in the current climate, each new data point from here on out brings the potential for a genuine case of regime shift.