Employers added 29,000 jobs in September and the unemployment rate rose to 4.2%, according to the Bureau of Labor Statistics September jobs report released October 2. The same release also cut 60,000 jobs from the prior two months. July went from a gain of 21,000 to a loss of 10,000, and August fell from 162,000 to 133,000.
The pain is uneven, as The College Investor’s breakdown of unemployment rates by education level shows.
The jobless rate has held between 4.1% and 4.3% since March, while the rate for Black workers rose to 7.0%. Average hourly earnings are up 3.0% over the past year, per BLS. Anyone facing job loss should check how unemployment benefits work in their state right away.
Would you like to save this?
We’ll email this article to you, so you can come back to it later!
Why It Matters
Downward revisions have become the norm, and its troubling for those watching the labor market. Of the 22 months from October 2024 through July 2026 with final monthly estimates, 15 were revised lower, and every month of 2025 finished below its first print, according to BLS revision data.
Net, those revisions erased 661,000 jobs that were reported at the time. The first print drives the headlines, but then every month those numbers disappear.
Employers have averaged just 45,000 new jobs a month over the past year, a thin market for recent grads. The College Investor has tracked the tough job market facing new college graduates and a nearly 20% drop in entry-level software jobs.
.tci-chart{
–bg:#ffffff; –ink:#1a1a1a; –ink-2:#555; –ink-3:#8a8a8a;
–grid:#e6e6e6; –zero:#b5b5b5;
–first:#a9c4e4; –final:#1f5fa8; –down:#c0392b; –up:#1f5fa8; –pending:#c9c9c9;
–line:#1f5fa8; –tip-bg:#1a1a1a; –tip-ink:#fff;
background:var(–bg);color:var(–ink);font-family:-apple-system,BlinkMacSystemFont,’Segoe UI’,Helvetica,Arial,sans-serif;box-sizing:border-box;}
.tci-chart *{box-sizing:border-box;}
@media (prefers-color-scheme: dark){ .tci-chart{
–bg:#121417; –ink:#f2f2f2; –ink-2:#b8b8b8; –ink-3:#8a8a8a;
–grid:#2a2d31; –zero:#5a5f66;
–first:#3d5a7d; –final:#6aa7e8; –down:#e8776a; –up:#6aa7e8; –pending:#4a4e54;
–line:#6aa7e8; –tip-bg:#f2f2f2; –tip-ink:#121417;
} }
.tci-chart{max-width:760px;margin:0 auto 28px;padding:16px;}
.tci-chart h3{font-size:18px;line-height:1.3;margin:0 0 4px;font-weight:600;}
.tci-chart p.sub{font-size:14px;color:var(–ink-2);margin:0 0 10px;}
.tci-chart .wrap{position:relative;}
.tci-chart svg{width:100%;height:auto;display:block;overflow:visible;}
.tci-chart .grid line{stroke:var(–grid);stroke-width:1;}
.tci-chart .zero{stroke:var(–zero);stroke-width:1;}
.tci-chart .axis text{fill:var(–ink-3);font-size:11px;font-variant-numeric:tabular-nums;}
.tci-chart .ptitle{fill:var(–ink-2);font-size:12px;font-weight:600;}
.tci-chart .series{fill:none;stroke:var(–line);stroke-width:2;stroke-linejoin:round;stroke-linecap:round;}
.tci-chart .dot{fill:var(–line);stroke:var(–bg);stroke-width:1.5;}
.tci-chart .lbl{fill:var(–ink);font-size:11px;font-weight:600;font-variant-numeric:tabular-nums;}
.tci-chart .hit{fill:transparent;cursor:pointer;}
.tci-chart .hl{fill:var(–ink);opacity:0;}
.tci-chart .tip{position:absolute;pointer-events:none;background:var(–tip-bg);color:var(–tip-ink);font-size:12px;line-height:1.4;padding:6px 8px;border-radius:4px;white-space:nowrap;opacity:0;transform:translate(-50%,-100%);transition:opacity .08s;}
.tci-chart .legend{display:flex;flex-wrap:wrap;gap:14px;font-size:12px;color:var(–ink-2);margin:0 0 6px;}
.tci-chart .legend span{display:inline-flex;align-items:center;gap:6px;}
.tci-chart .sw{width:11px;height:11px;border-radius:2px;display:inline-block;}
.tci-chart .stats{display:flex;flex-wrap:wrap;gap:10px 28px;margin:4px 0 12px;}
.tci-chart .stat b{display:block;font-size:22px;font-weight:600;font-variant-numeric:tabular-nums;}
.tci-chart .stat span{font-size:12px;color:var(–ink-2);}
.tci-chart .foot{font-size:12px;color:var(–ink-3);margin-top:8px;line-height:1.45;}
.tci-chart details{margin-top:8px;font-size:13px;}
.tci-chart summary{cursor:pointer;color:var(–ink-2);}
.tci-chart .tbl{overflow-x:auto;}
.tci-chart table{border-collapse:collapse;margin-top:8px;font-variant-numeric:tabular-nums;font-size:12px;}
.tci-chart td, .tci-chart th{padding:3px 14px 3px 0;text-align:right;border-bottom:1px solid var(–grid);}
.tci-chart td:first-child, .tci-chart th:first-child{text-align:left;}
.tci-chart th{color:var(–ink-2);font-weight:600;}
Monthly Job Gains: First Report vs. Final Estimate, Oct. 2024–Sept. 2026
Change in total nonfarm payrolls (thousands), as first reported and after the two scheduled monthly revisions
Final monthly estimate
Revised down
Revised up
Revisions still pending
View data table
U.S. Unemployment Rate, Sept. 2024–Sept. 2026
Seasonally adjusted, percent. The unemployment rate comes from the household survey and is not revised month to month.
(function(){
var NS=”http://www.w3.org/2000/svg”;
function el(tag,attrs,parent){var e=document.createElementNS(NS,tag);for(var k in attrs)e.setAttribute(k,attrs[k]);if(parent)parent.appendChild(e);return e;}
function fmt(v){return (v>0?”+”:v0?”+”+v:v;
});
[-150,-100,-50,0,50].forEach(function(v){
el(“line”,{x1:L,x2:L+W,y1:y2(v),y2:y2(v),”class”:v===0?”zero”:””},g);
var t=el(“text”,{x:L-6,y:y2(v)+4,”text-anchor”:”end”},ax);t.textContent=v>0?”+”+v:v;
});
var pt=el(“text”,{x:L,y:T2-10,”class”:”ptitle”},svg);pt.textContent=”Revision: final minus first report (thousands)”;
var hlg=el(“g”,{},svg), bars=el(“g”,{},svg);
D.forEach(function(d,i){
var cx=L+step*i+step/2;
var hl=el(“rect”,{x:cx-step/2,y:T1,width:step,height:T2+H2-T1,”class”:”hl”},hlg);
el(“rect”,{x:cx-bw,y:Math.min(y1(d[1]),y1(0)),width:bw,height:Math.abs(y1(d[1])-y1(0)),fill:”var(–first)”},bars);
if(d[2]!==null) el(“rect”,{x:cx,y:Math.min(y1(d[2]),y1(0)),width:bw,height:Math.abs(y1(d[2])-y1(0)),fill:d[3]===”p”?”var(–pending)”:”var(–final)”},bars);
if(d[2]!==null){
var r=d[2]-d[1], col=d[3]===”p”?”var(–pending)”:(r<0?"var(–down)":"var(–up)");
el("rect",{x:cx-step*0.32,y:Math.min(y2(r),y2(0)),width:step*0.64,height:Math.max(1,Math.abs(y2(r)-y2(0))),fill:col},bars);
}
if(i%3===0||i===n-1){var t=el("text",{x:cx,y:T2+H2+16,"text-anchor":"middle"},ax);t.textContent=d[0];}
var hit=el("rect",{x:cx-step/2,y:T1,width:step,height:T2+H2-T1,"class":"hit"},svg);
function show(){
hl.style.opacity=.05;
var s="“+d[0].replace(“‘”,”20″)+”
First report: “+fmt(d[1]);
if(d[3]===”n”) s+=”
Revisions pending”;
else s+=”
“+(d[3]===”p”?”Second estimate: “:”Final estimate: “)+fmt(d[2])+”
Revision: “+fmt(d[2]-d[1])+(d[3]===”p”?” (so far)”:””);
tip.innerHTML=s;
var bb=svg.getBoundingClientRect(), sc=bb.width/728, x=cx*sc;
x=Math.max(70,Math.min(bb.width-70,x));
tip.style.left=x+”px”;tip.style.top=(T1*sc+4)+”px”;tip.style.transform=”translate(-50%,0)”;tip.style.opacity=1;
}
function hide(){hl.style.opacity=0;tip.style.opacity=0;}
hit.addEventListener(“mouseenter”,show);hit.addEventListener(“mouseleave”,hide);
hit.addEventListener(“touchstart”,function(e){show();},{passive:true});
});
var tb=document.getElementById(“t1″);
tb.innerHTML=”
return “
}).join(“”);
// ———- Chart 2 ———-
var U=[[“Sep ’24”,4.1],[“Oct ’24”,4.1],[“Nov ’24”,4.2],[“Dec ’24”,4.1],[“Jan ’25”,4.0],[“Feb ’25”,4.2],[“Mar ’25”,4.2],[“Apr ’25”,4.2],[“May ’25”,4.3],[“Jun ’25”,4.1],[“Jul ’25”,4.3],[“Aug ’25”,4.3],[“Sep ’25”,4.4],[“Oct ’25”,null],[“Nov ’25”,4.5],[“Dec ’25”,4.4],[“Jan ’26”,4.3],[“Feb ’26”,4.4],[“Mar ’26”,4.3],[“Apr ’26”,4.3],[“May ’26”,4.3],[“Jun ’26”,4.2],[“Jul ’26”,4.1],[“Aug ’26”,4.1],[“Sep ’26”,4.2]];
var s2=document.getElementById(“ur”), tip2=document.getElementById(“tip2”);
var L2=40,R2=14,W2=728-L2-R2,T=16,H=240, m=U.length, st=W2/(m-1);
var yy=function(v){return T+(4.7-v)/(4.7-3.8)*H;}, xx=function(i){return L2+st*i;};
var g2=el(“g”,{“class”:”grid”},s2), a2=el(“g”,{“class”:”axis”},s2);
[3.8,4.0,4.2,4.4,4.6].forEach(function(v){
el(“line”,{x1:L2,x2:L2+W2,y1:yy(v),y2:yy(v)},g2);
var t=el(“text”,{x:L2-6,y:yy(v)+4,”text-anchor”:”end”},a2);t.textContent=v.toFixed(1)+”%”;
});
var segs=[[]];U.forEach(function(u,i){if(u[1]===null){segs.push([]);}else segs[segs.length-1].push(xx(i)+”,”+yy(u[1]));});
segs.forEach(function(sg){if(sg.length)el(“polyline”,{points:sg.join(” “),”class”:”series”},s2);});
// dashed bridge over missing Oct 2025
el(“line”,{x1:xx(12),y1:yy(4.4),x2:xx(14),y2:yy(4.5),stroke:”var(–line)”,”stroke-width”:1.5,”stroke-dasharray”:”3 4″},s2);
var cross=el(“line”,{y1:T,y2:T+H,stroke:”var(–ink-3)”,”stroke-width”:1,opacity:0},s2);
U.forEach(function(u,i){
if(u[1]!==null) el(“circle”,{cx:xx(i),cy:yy(u[1]),r:3,”class”:”dot”},s2);
if(i%3===0){var t=el(“text”,{x:xx(i),y:T+H+18,”text-anchor”:”middle”},a2);t.textContent=u[0];}
var h=el(“rect”,{x:xx(i)-st/2,y:T,width:st,height:H,”class”:”hit”},s2);
function show(){
cross.setAttribute(“x1”,xx(i));cross.setAttribute(“x2”,xx(i));cross.setAttribute(“opacity”,.5);
tip2.innerHTML=”“+u[0].replace(“‘”,”20″)+”
“+(u[1]===null?”Not collected (shutdown)”:u[1].toFixed(1)+”%”);
var bb=s2.getBoundingClientRect(), sc=bb.width/728, x=Math.max(60,Math.min(bb.width-60,xx(i)*sc));
tip2.style.left=x+”px”;tip2.style.top=((u[1]===null?yy(4.45):yy(u[1]))*sc-10)+”px”;tip2.style.opacity=1;
}
h.addEventListener(“mouseenter”,show);h.addEventListener(“mouseleave”,function(){cross.setAttribute(“opacity”,0);tip2.style.opacity=0;});
h.addEventListener(“touchstart”,show,{passive:true});
});
[[4,”4.0%”,1],[14,”4.5%”,-1],[24,”4.2%”,-1]].forEach(function(p){
var t=el(“text”,{x:xx(p[0]),y:yy(U[p[0]][1])+(p[2]>0?18:-10),”text-anchor”:p[0]===24?”end”:”middle”,”class”:”lbl”},s2);t.textContent=p[1];
});
})();
Why Revisions Keep Going Down
It’s hard to miss the trend of revisions downward. Here’s why this keeps happening.
- Late responses: In 2024, BLS had data from about 60% of surveyed employers when it published first estimates, rising to about 91% by the third estimate, per the Congressional Research Service.
- Falling participation: CES response rates dropped from about 58% before 2020 to 43% in 2024, the same CRS report found. In August 2025, nearly half of the monthly revision came from late government payroll data, mostly in state and local education.
- Turning points: Cleveland Fed researchers found revisions near turning points “tend to be procyclical,” moving in the direction the economy is already heading.
- Annual benchmarks: The March 2025 benchmark removed 898,000 jobs. BLS adjusted its birth-death model in February 2026, and the preliminary March 2026 benchmark was just -79,000.
The idea that BLS inflates first prints and quietly walks them back gained traction after President Trump fired BLS Commissioner Erika McEntarfer on August 1, 2025, alleging manipulation without evidence. However, that has not been proven with any compelling data.
Revisions happen, and they can move massively in both directions. This same process produced upward revisions of 437,000 and 389,000 for November and December 2021 and a 506,000-job benchmark increase in 2022. The Yale Budget Lab found nonresponse, not the birth-death model, drove most of the 2024-2025 issues.
However, questions about federal data reliability extend beyond BLS, as we’ve already covered with missing Education Department student loan reports.
How This Connects
A soft job market lands on student loan borrowers at a bad time, with 9.3 million federal borrowers now in default and wage garnishment restarting.
Student loan borrowers who lose work should take action before they fall into delinquency and default. The best action to take is to use this unemployment to enroll in an income-driven repayment plan, which could potentially allow you to have a very low or even $0 monthly payment.
The second best option is to request an unemployment deferment instead of falling behind.
What’s Next
The October report will revise August a final time and September for the first time. BLS will publish the final March 2026 benchmark with January data in early 2027.
Editor: Colin Graves
The post September Jobs Report: 29,000 Jobs Added As BLS Cuts 60,000 From July And August appeared first on The College Investor.