Meta tags:
description=15 posts published by Lars Syll during February 2026;

Headings (most frequently used words):

the, why, of, recent, comments, lars, syll, katalys, nya, chefsekonom, table, fallacy, conflating, prediction, and, causation, minsky, still, matters, trump, tariffs, laughing, matter, vienna, song, für, friedrich, merz, kritiken, av, svantesson, totalt, missriktad, are, ceos, paid, so, much, finanspolitiska, rådet, drar, helt, fel, slutsats, dangers, insufficiently, deep, explanations, ivory, coast, how, to, make, full, employment, compatible, with, fiscal, stability, causal, inference, in, presence, heterogeneity, econometrics, second, best, explanatory, practice, we, all, care, about, inequality, posts, policy, reading, list, categories, archives,

Text of the page (most frequently used words):
the (356), and (164), that (103), att (85), och (74), not (71), are (60), det (52), share (47), new (47), opens (45), window (45), with (44), for (44), som (38), this (35), have (33), feb (31), inte (31), ett (31), för (31), bluesky (30), facebook (30), may (28), they (28), #causal (28), den (27), can (27), 2026 (23), more (23), independent (21), effect (21), than (20), financial (20), med (20), till (20), jan (19), economic (19), one (19), when (19), comments (18), other (18), what (18), income (18), between (18), but (17), about (17), variables (17), from (17), has (17), mar (16), apr (16), jun (16), posted (16), men (16), kan (16), jul (15), aug (15), sep (15), oct (15), nov (15), dec (15), school (15), har (15), 2021 (14), 2023 (14), theory (14), arbetslöshet (14), will (14), full (14), would (14), their (14), effects (14), pupils (14), wealth (14), 2025 (13), economics (13), often (13), without (13), sweden (13), debt (13), 2012 (12), 2013 (12), 2014 (12), 2015 (12), 2016 (12), 2017 (12), 2018 (12), 2019 (12), 2020 (12), 2022 (12), 2024 (12), econometrics (12), education (12), off (12), problem (12), there (12), vara (12), sig (12), lars (11), lönebildning (11), does (11), those (11), because (11), example (11), all (11), variable (11), problems (11), schools (11), finanspolitiska (11), måste (11), man (11), crises (11), 2011 (10), how (10), inequality (10), some (10), why (10), control (10), based (10), which (10), even (10), natural (10), eller (10), när (10), society (9), better (9), should (9), also (9), fact (9), analysis (9), over (9), however (9), factors (9), where (9), risks (9), economy (9), market (9), time (8), only (8), policy (8), ekonomiska (8), many (8), who (8), means (8), variation (8), these (8), question (8), regression (8), into (8), heterogeneity (8), detta (8), report (7), like (7), sverige (7), most (7), group (7), different (7), while (7), such (7), statistical (7), mainstream (7), causes (7), knowledge (7), use (7), explanatory (7), whether (7), requires (7), then (7), system (7), result (7), relationship (7), crisis (7), its (7), underlying (7), assets (7), rådet (7), comment (6), account (6), statistics (6), arbetslösheten (6), model (6), research (6), rather (6), themselves (6), social (6), given (6), economists (6), each (6), 000 (6), keynes (6), second (6), must (6), information (6), difficult (6), out (6), truly (6), randomised (6), experiments (6), sometimes (6), know (6), long (6), again (6), our (6), parents (6), among (6), high (6), both (6), public (6), was (6), through (6), distribution (6), ska (6), finanspolitik (6), ekonomin (6), sysselsättning (6), mål (6), investeringar (6), inget (6), vienna (6), minsky (6), already (5), now (5), arbetslöshetssiffran (5), styr (5), recent (5), first (5), feldstein (5), been (5), thus (5), them (5), just (5), suppose (5), reasons (5), relevant (5), methods (5), answer (5), empirical (5), another (5), measure (5), real (5), way (5), background (5), choice (5), situation (5), could (5), swedish (5), approach (5), instability (5), yours (5), employment (5), stability (5), fiscal (5), unequal (5), demand (5), risk (5), ramverk (5), mycket (5), politik (5), ramverket (5), sin (5), minska (5), regeringar (5), generationer (5), staten (5), uppnå (5), skulden (5), eftersom (5), utan (5), since (5), future (5), earnings (5), syll (4), varia (4), politics (4), fred (4), especially (4), being (4), same (4), statsskulden (4), feel (4), worse (4), always (4), hold (4), own (4), let (4), still (4), away (4), much (4), care (4), itself (4), processes (4), causation (4), instruments (4), possible (4), best (4), basic (4), very (4), things (4), controlling (4), conclusion (4), constant (4), taken (4), power (4), levels (4), here (4), relationships (4), become (4), important (4), studies (4), precisely (4), reality (4), therefore (4), make (4), method (4), economist (4), significant (4), able (4), complete (4), results (4), municipal (4), socio (4), status (4), need (4), recurring (4), misleading (4), had (4), estimated (4), lies (4), completely (4), higher (4), clear (4), associated (4), issues (4), households (4), marginal (4), total (4), questions (4), assumptions (4), markets (4), också (4), minst (4), fel (4), grunden (4), finanspolitiken (4), ökad (4), bra (4), oss (4), hålla (4), bnp (4), väl (4), dem (4), sätt (4), bara (4), två (4), skuld (4), skulle (4), pengar (4), säger (4), helt (4), slut (4), concentration (4), direct (4), according (4), year (4), history (4), damberg (4), expectations (4), borrowers (4), stochastic (4), fallacy (4), table (4), occupation (4), science (3), list (3), nothing (3), politiken (3), merely (3), people (3), values (3), hence (3), large (3), well (3), were (3), united (3), states (3), his (3), conclusions (3), greater (3), yet (3), might (3), upon (3), reason (3), less (3), behind (3), part (3), never (3), relations (3), explanations (3), deep (3), matter (3), practice (3), explain (3), usually (3), ask (3), ourselves (3), really (3), draw (3), forces (3), provide (3), dependent (3), theoretical (3), modern (3), any (3), far (3), kind (3), squared (3), find (3), ability (3), graph (3), traditional (3), under (3), regarding (3), population (3), random (3), give (3), consequences (3), look (3), institutions (3), exogenous (3), john (3), maynard (3), further (3), terms (3), addition (3), attended (3), obviously (3), unobservable (3), assume (3), experiment (3), exists (3), affects (3), selection (3), mechanism (3), cannot (3), end (3), logic (3), role (3), estimates (3), mechanisms (3), immigrant (3), take (3), functional (3), necessarily (3), mean (3), differences (3), years (3), etc (3), presence (3), degree (3), issue (3), several (3), models (3), inference (3), budget (3), macroeconomic (3), outcomes (3), policies (3), abba (3), lerner (3), state (3), deficits (3), though (3), interest (3), largely (3), equal (3), reducing (3), consume (3), top (3), overall (3), credit (3), central (3), estimate (3), events (3), självklart (3), rådets (3), borde (3), länge (3), sedan (3), sagt (3), vår (3), faktum (3), regelverk (3), varit (3), skapa (3), mer (3), rapport (3), mot (3), hur (3), regeringen (3), framtida (3), ökade (3), kommande (3), innebär (3), balansmål (3), idag (3), budgetunderskott (3), procent (3), även (3), offentlig (3), använda (3), vill (3), var (3), bör (3), sina (3), makroekonomiska (3), prisstabilitet (3), stora (3), statsskuld (3), denna (3), börda (3), rätt (3), skatter (3), räntan (3), sparande (3), oxfam (3), extreme (3), deliberate (3), tax (3), systematically (3), world (3), poverty (3), productivity (3), safety (3), coefficient (3), balance (3), normal (3), borrower (3), banks (3), required (2), log (2), sign (2), subscribed (2), subscribe (2), wordpress (2), com (2), subscribers (2), blog (2), wiklund (2), milch (2), follow (2), ones (2), limited (2), randomisation (2), contributions (2), randomness (2), credible (2), credibility (2), problemet (2), lesson (2), precision (2), student (2), next (2), illustrate (2), received (2), cents (2), justice (2), hurt (2), compare (2), receive (2), goods (2), services (2), tangible (2), worth (2), difference (2), viewed (2), wrong (2), bit (2), fairy (2), gives (2), participants (2), welfare (2), refuse (2), quarter (2), others (2), throw (2), members (2), reveal (2), say (2), obtaining (2), tend (2), subject (2), argued (2), book (2), primarily (2), stanley (2), lieberson (2), randomly (2), determined (2), controlled (2), making (2), econometric (2), genuine (2), confounders (2), identifying (2), simple (2), observed (2), understanding (2), event (2), context (2), certain (2), place (2), absence (2), fundamentally (2), considering (2), impact (2), kincaid (2), self (2), evident (2), variances (2), shows (2), measures (2), influence (2), variance (2), instead (2), sample (2), useful (2), causality (2), criticism (2), comes (2), evaluating (2), definitely (2), assumed (2), directly (2), touched (2), included (2), treated (2), rise (2), outcome (2), extent (2), turn (2), impossible (2), perceived (2), right (2), having (2), correct (2), three (2), enough (2), factor (2), quantitative (2), applicable (2), neither (2), discovery (2), nor (2), familiar (2), called (2), attending (2), her (2), she (2), confounder (2), via (2), work (2), affect (2), children (2), choose (2), attend (2), once (2), affected (2), type (2), ideal (2), distributed (2), before (2), help (2), active (2), try (2), common (2), analyses (2), beyond (2), educational (2), within (2), category (2), perhaps (2), immeasurable (2), perform (2), play (2), significantly (2), groups (2), focus (2), respects (2), connected (2), ethnicity (2), comparison (2), term (2), homogeneity (2), becomes (2), researchers (2), lead (2), pearl (2), mostly (2), tool (2), graphs (2), additivity (2), interaction (2), solve (2), necessary (2), balanced (2), austerity (2), unemployment (2), price (2), capacity (2), prices (2), borrowing (2), invested (2), government (2), bonds (2), payments (2), pay (2), taxes (2), net (2), reduce (2), run (2), spending (2), progressive (2), taxation (2), collective (2), bargaining (2), asset (2), lower (2), propensity (2), source (2), compatible (2), weakens (2), household (2), major (2), larger (2), consumption (2), reliance (2), aggregate (2), post (2), keynesian (2), wealthy (2), hands (2), spent (2), ivory (2), coast (2), treatment (2), works (2), deeper (2), frequently (2), identification (2), design (2), course (2), essential (2), reaching (2), rosenzweig (2), wolpin (2), number (2), needed (2), small (2), dangers (2), insufficiently (2), vårt (2), land (2), tanke (2), oavsett (2), överens (2), nuvarande (2), öppna (2), går (2), alltid (2), förfelat (2), principer (2), snarare (2), finansminister (2), själv (2), låna (2), framför (2), kritik (2), vilket (2), leda (2), drar (2), missriktad (2), sunda (2), statsfinanser (2), instabila (2), priser (2), överges (2), kristider (2), mellan (2), efterfrågestimulerande (2), infrastruktur (2), sikt (2), ekonomi (2), utgiftstak (2), kommer (2), onödan (2), förändras (2), redan (2), talet (2), underskott (2), ansvarsfull (2), tillbaka (2), fart (2), läge (2), dagens (2), perspektiv (2), kanske (2), fungerar (2), behöver (2), vid (2), balans (2), målen (2), mindre (2), där (2), står (2), kvar (2), enligt (2), runt (2), ligga (2), hög (2), allt (2), både (2), andra (2), varken (2), dålig (2), övergripande (2), vad (2), heligt (2), balanserad (2), medel (2), konsoliderade (2), maastrichtskulden (2), statsbudgeten (2), stället (2), värna (2), dock (2), genom (2), effekter (2), faktiskt (2), göra (2), våra (2), barn (2), barnbarn (2), betalar (2), argumentet (2), par (2), beror (2), infrastruktursatsningar (2), något (2), finansieras (2), samhället (2), håller (2), obligationer (2), ränta (2), medan (2), betala (2), balanserar (2), låg (2), politiskt (2), lag (2), ökar (2), privata (2), från (2), ingen (2), utgifter (2), valutor (2), eget (2), värde (2), silver (2), gör (2), blir (2), nya (2), ekonomisk (2), hade (2), ske (2), disproportionate (2), political (2), transparency (2), prevent (2), gap (2), consequence (2), decades (2), burdens (2), capital (2), actively (2), highest (2), ranks (2), country (2), perception (2), societies (2), evidence (2), individuals (2), 700 (2), unable (2), deprivation (2), combined (2), bottom (2), elite (2), trickle (2), down (2), feed (2), pass (2), trying (2), today (2), individual (2), little (2), ceos (2), times (2), paid (2), level (2), neoliberal (2), rich (2), 1980s (2), annat (2), hål (2), huvudet (2), daniel (2), suhonen (2), talar (2), möjliggöra (2), ännu (2), socialdemokraterna (2), kritiken (2), totalt (2), mikael (2), svantesson (2), pengarna (2), han (2), song (2), für (2), friedrich (2), merz (2), karlsbad (2), you (2), highly (2), ago (2), series (2), university (2), pleasure (2), studying (2), great (2), hyman (2), fundamental (2), debts (2), insignificant (2), regulations (2), minimise (2), speculation (2), leads (2), endogenous (2), created (2), distribute (2), coin (2), heads (2), tails (2), bubbles (2), securitisation (2), ratings (2), representative (2), sheets (2), liquidity (2), creates (2), during (2), upswing (2), phase (2), margins (2), structure (2), conditioning (2), indirect (2), occupational (2), conflating (2), prediction (2), katalys (2), website, name, email, write, loading, collapse, bar, manage, subscriptions, view, site, reader, content, privacy, join, 606, feeds, entries, 112, archives, 707, 542, methodology, 998, 391, 301, 168, categories, reading, fredtorssander, torssander, netiquette, anonymous, pseudo, argumentations, abusive, language, irrelevant, links, please, remember, professor, leaves, respond, reification, guarantee, nancy, cartwright, philosophy, representativeness, revolution, välförtjänt, bottenrekord, audioholic, summer, stein, paradox, unbiasedness, love, secret, finding, late, stuff, efter, studenten, personal, risksamhällets, konturer, posts, page, branko, milanovic, story, suggests, feeling, propriety, peers, acquire, recognition, expression, particularly, unjustified, unclear, slight, consider, martin, 1999, heeled, gathered, conference, journal, asks, apparently, modify, variously, previous, everybody, dare, suggest, quite, accept, leave, room, disgust, unfavorably, unfathomable, worthy, 1000th, amount, speculate, lie, extravagant, largesse, foundation, variations, brought, tie, actions, structures, detected, overly, enamoured, axiomatic, deductive, modelling, forget, accounting, bring, demands, acquaintance, intricate, contexts, fabric, seen, describing, patterns, associations, correlations, suggestions, forgetting, continue, stuck, treatise, probability, critique, strikes, heart, interprets, want, assigned, covariates, inferences, answering, moreover, questionable, wit, vital, least, working, hypothesis, causing, per, magnitude, accounts, almost, universal, criterion, forthcoming, undermined, assuming, explicit, justification, gain, amenable, supplies, count, harold, fare, explaining, influences, silent, similarly, size, varies, none, arbitrary, noncausal, epistemic, roles, measuring, despite, certainly, baby, bathwater, adopting, healthy, scepticism, caution, judging, concerning, establish, recommended, hubert, blalock, left, outside, actually, operate, produce, nonrandom, confounded, consideration, overlook, aspect, seldom, determines, something, observational, quasi, anywhere, near, rules, anticipated, thinking, multiple, correlation, essentially, depends, furnished, goes, vera, causa, discover, relative, importance, beforehand, indubitably, giving, qualitative, pupil, partly, schoolmates, similar, sense, benefits, environment, happen, elimination, unmeasurable, attributes, send, differ, function, fully, adequate, substitutes, assignment, academic, performance, introduction, probably, existing, simulation, proceed, causally, parts, reliably, simply, looks, unfortunately, confronted, overlooked, get, procedure, influencing, interested, simulate, resembles, match, comparable, hope, obtain, pure, catch, hidden, attitude, motivation, select, believe, debate, argument, segregation, access, mask, decisive, held, theorem, arise, creating, aware, support, compensatory, counteract, prejudices, presumed, teaching, mentioned, consequently, coincide, specification, appear, ethnic, backgrounds, differently, originate, instance, alternative, basis, begin, general, equivalent, lump, together, taking, pointless, politicians, asked, friskolor, charter, test, multitude, known, observable, housing, field, advocates, emphasise, advantages, flexibility, disadvantages, lack, clarity, characteristics, present, specified, create, judea, puts, forward, compelling, popular, theoretic, preferred, non, parametric, arrows, nodes, replace, specifications, equations, sacred, reduction, irrespective, sound, finances, budgets, worsening, climate, reconsidered, proponents, monetary, responsibility, promote, powers, accordingly, running, temporary, accumulating, inherently, good, bad, instrument, achieving, two, overarching, objectives, stable, fear, financed, domestically, foreign, benefit, productively, ultimately, finance, perspective, whole, burden, cancels, creditors, taxpayers, low, rates, generating, inflationary, pressure, discipline, permanent, stimulus, implication, objective, limit, accumulation, triggering, stagnation, cutting, deficient, deepens, durable, solution, strong, building, raise, strengthening, contrast, concentrated, maintain, output, lean, heavily, asymmetry, evenly, national, flows, structural, rises, reach, smaller, sustained, propensities, link, heterodox, sidelined, debates, extra, euro, poorer, richer, likely, saved, typically, delivers, decomposition, captures, specific, compliers, clarifying, distinct, pathways, operates, learn, limits, development, external, validity, separable, channels, remaining, superficial, complementary, unpack, quantify, analysed, using, instrumental, attractive, offer, recover, trials, infeasible, endogeneity, threatens, careful, planning, ultimate, meaningful, citizens, scholarly, community, cases, primary, scholars, construct, sufficiently, clever, generalise, broader, populations, shifts, attention, substantive, towards, methodological, refinement, sake, discuss, serious, notable, address, rely, strategies, formulate, ways, yield, exact, precise, answers, mark, kenneth, sufficient, eliminate, necessity, imposing, auxiliary, supplementary, measurement, valuable, ever, evidently, potential, strictly, experimental, relies, exclusively, constrained, provided, nature, preferences, technologies, prospect, granska, förs, frågan, egen, verksamhet, granskas, ord, problematiserar, kritiskt, analyserar, skrotats, regeringens, undermålig, möjligheten, takt, förd, respekterar, ramverks, ofta, skäl, nog, ifrågasätta, citera, gud, riksdagen, beslutar, ändrar, tycka, anmärkningsvärt, regering, efterleva, väljer, samtidigt, skatten, sänker, årets, hittills, allvarligaste, förhåller, ramverkets, landar, främst, fördömande, skjuter, finansieringen, försvarsutgifterna, sägs, reformutrymmet, bördor, därför, slutsatsen, genomgång, förda, åtstramningar, balanserade, budgetar, fortskridande, klimatkris, ont, goda, bruksanvisningar, fastna, stagnerande, konsumtionens, skylla, restriktiva, kreditgivningens, karybdis, krävs, rejäla, satsningar, ordentlig, arbetslöshetsersättning, generöst, socialförsäkringssystem, offensiv, aktivt, bidra, riskerna, kostsamma, systemkriser, egenstickade, tvångströjor, form, finanspolitiskt, tyvärr, förlänga, rådande, kriser, tider, förr, nödvändigt, förödande, dåligt, lärde, 1930, nödvändig, del, jobb, tillväxt, bli, tvungna, acceptera, storleksordningen, sådant, bättre, rakryggat, öppet, deklarera, övergivande, sett, tidigare, stärkt, solen, skiner, mestadels, väsentligt, sämre, regnar, motvind, uppförsbacke, överleva, kräftgången, kontraproduktivt, låtsas, fast, utgifts, internationell, jämförelse, framstår, exceptionellt, restriktivt, betydligt, ambitiöst, exempel, stabilitets, tillväxtpakt, ideologiskt, passat, fot, strumpa, velat, sektor, verkligen, kunna, medlet, adekvat, helig, slaktas, ändalykten, krita, vem, helst, räkna, befinner, extraordinär, extraordinära, åtgärder, bidrar, förvärra, stampar, ligger, över, scb, senaste, skattning, begåvats, ben, varför, hoppa, konjunkturen, riktigt, vända, fortsätter, skrämmande, nivå, svåra, blivit, tydligare, penning, räntepolitiken, längre, verkningsfull, liksom, mmt, förespråkare, uppenbart, förmågan, främja, stabilt, prisnivå, krafter, underbalansera, budgeten, offentliga, påverkar, svenska, utlandsskulden, historiskt, låga, framgår, hör, länder, allra, lägst, skuldsättning, utmaningar, inför, aggressivare, miltärmakt, öst, fortsatt, tal, ansvar, oansvarigt, statsfinanserna, rege, ring, samhällets, framtid, nere, bruttoskulden, medellång, idén, leder, sanningen, kallade, generationsövergripande, normalfallet, verklig, används, lämpligt, nettovinnare, naturligtvis, finnas, oönskade, negativa, fördelningsmässiga, oftast, dessa, betalningar, göras, återkommit, flera, gånger, lånefinansiering, utgöra, detsamma, hörde, diskussionerna, hetast, underskottsmål, orättvist, belasta, vare, försvarsutgifter, vanligtvis, frukta, särskilt, inom, landet, utländska, lån, nytta, investeras, vissa, medlemmar, tjänar, slutligen, nettoförlust, helhet, grupperna, emitterar, nödvändigtvis, starkt, inflationstryck, storleken, 1940, argumenterade, teori, funktionell, nivån, vilka, spendera, betyder, förbehåll, beslut, ekonomer, egentligen, tänker, variant, says, lika, investeringarna, crowding, bokföringsmässig, identitet, ekonomiskt, ointressant, planerat, avviker, varandra, anpassar, efterhand, huruvida, lyckas, misslyckas, ingenting, ekonomins, värld, behövs, förutom, når, nollgräns, zlb, normala, omständigheter, räcker, bedriva, penningpolitik, centralbankerna, justera, inflation, budgetunderskottet, tenderar, tränga, räntorna, stiger, alla, moderna, baserade, fiat, intrinsikalt, guld, ger, enkelt, skatt, möjligt, driva, slags, monopol, aldrig, statsutgifter, därmed, verkliga, motorn, statslån, sekundära, verktyg, depression, lösningen, åtstramning, stort, fiatpengar, välgrundat, teoretiskt, tackla, finansieringsfrågan, föreslagit, skattefinansieringen, kraftigt, höjd, progression, inkomstskatterna, kapitalskatter, särskilda, övervinstskatter, jag, kunnat, köpa, förslaget, lär, skattesänkarregering, sitter, politiska, rodret, presenterades, häromdagen, föreslås, istället, kraftfulla, aviserade, försvarssatsningarna, uppemot, 300, miljarder, kronor, höjda, omprioriteringar, årliga, försvaret, rekommendera, slutsats, rightly, contends, poses, existential, threat, extends, corrodes, foundational, principles, democracy, translates, eroding, institutional, trust, undermining, principle, citizenship, calls, minor, adjustments, shift, establishment, comprehensive, register, radical, lobbying, aggressively, concludes, celebrated, sustaining, urgent, intervention, irreversible, unravelling, contract, critically, frames, widening, phenomenon, regressive, reforms, inheritance, increasing, engineered, last, ubs, global, sixth, places, ahead, countries, commonly, strikingly, seventh, index, underscores, severity, challenges, widespread, outlier, defined, equality, globally, overwhelming, alarming, surge, 120, single, expands, impoverished, approximately, concurrently, afford, energy, measured, inability, heat, home, tripled, contrasts, violently, fortunes, apex, billionaires, surpasses, poorest, million, grew, alone, earners, disposable, roughly, damning, indictment

Text of the page (random words):
bility this new tool gives us there are also clear risks and disadvantages associated with the use of causal graphs the lack of clarity regarding whether additivity interaction or other variable and relationship characteristics are present and how they if so are specified can sometimes create more problems than they solve many of the problems just as with regression analyses are connected to the presence and degree of heterogeneity let me take an example from the field of educational research to illustrate the issue in recent years a recurring question that both politicians and researchers in sweden have asked themselves is whether our friskolor independent schools charter schools lead to higher knowledge levels and test results among the pupils to be able to answer this in reality very difficult causal question we need knowledge of a multitude of known observable variables and background factors parents income and education ethnicity housing etc etc in addition also factors that we know are important but are unobservable and or more or less immeasurable the problems begin already when we ask ourselves what lies behind the general term independent school not all independent schools are equivalent homogeneity we know that there are often large differences between them heterogeneity to then lump them all together and try to answer the causal question without taking these differences into account often becomes pointless and sometimes completely misleading another problem is that another type of heterogeneity which has to do with the specification of the functional relationships can appear suppose the independent school effect is connected to for example ethnicity and that pupils with a swedish background perform better than pupils with an immigrant background this does not necessarily mean that pupils with different ethnic backgrounds are in themselves affected differently by attending an independent school the effect could rather originate for instance from the fact that the alternative municipal schools that immigrant pupils could have attended were worse than those the swedish pupils could have attended if these differences in the basis for comparison are not taken into account the estimated independent school effects become misleading further heterogeneity problems arise if the mechanisms that are active in creating the independent school effect look significantly different for different groups of pupils independent schools with a focus on immigrant groups could for example be more aware of the need to support these pupils and take compensatory measures to counteract prejudices and the like in addition to the effects of the presumed better teaching in other respects at independent schools the effects for this category of pupils are also an effect of the mentioned heterogeneity and will consequently not coincide with that for the other group of pupils unfortunately the problems do not end here we are also confronted with a difficult and often overlooked selection problem when we try to get an answer to the causal question about the effects of independent schools a common procedure in regression analyses is to hold constant or control for influencing factors beyond those we are primarily interested in when it comes to independent schools a common control variable is parents income or educational background the logic is that in this way we can simulate an ideal situation that resembles as much as possible a randomised experiment where we only compare match pupils to parents with comparable education or income and in this way hope to obtain a better measure of the pure independent school effect the catch here is that within each income and education category there may be hidden further sometimes unobservable and perhaps immeasurable heterogeneity that has to do with for example attitude and motivation and which means that some pupils tend to choose select to attend independent schools because they believe they know that they will perform better there than at municipal schools in the independent school debate a recurring argument regarding segregation effects is that pupils with parents of high socio economic status have better access to information about the effects of school choice than other pupils the income or education variable can thus in fact mask other factors that can sometimes play a more decisive role than they do the estimates of the independent school effect can therefore again be misleading and sometimes even more misleading than if we had not held constant for any control variable at all cf the second best theorem in welfare economic theory controlling for possible confounders is thus not always obviously the right way to go if the very relationship between independent school x and academic performance y is affected by the introduction of the control variable socio economic status w this is probably a result of some type of relationship existing between x and w this also means that we do not have an ideal experiment simulation because there are obviously factors that affect y and that are not randomly distributed randomised before we can proceed we must then ask ourselves why the relationship in question exists to be able to causally explain the relationship between x and y we need to know more about how w affects the choice of x among other things we may then find that there is a difference in the choice of x between different parts of the group with high socio economic status w without knowledge of this selection mechanism we cannot reliably measure x s effect on y the randomised explanatory model is simply not applicable without knowledge of why a relationship exists and what it looks like between x and w controlling does not help us because it does not account for the active selection mechanism in addition to the problems touched upon here we have other long familiar problems the so called context or group effect for a pupil attending an independent school the results may partly be an effect of the fact that her schoolmates have a similar background and that she therefore in some sense benefits from her environment which would not happen if she attended a municipal school again means that confounder elimination via control variables does not obviously work when there is a relationship between the control variable and unmeasurable or difficult to measure unobservable attributes that themselves affect the dependent variable in our school example one can assume that the parents with a certain socio economic status who send their children to independent schools differ from the same group of parents who choose to let their children attend a municipal school the control variables do not once again function as fully adequate substitutes for a real experiment s randomised assignment am i right in thinking that the method of multiple correlation analysis essentially depends on the economist having furnished not merely a list of the significant causes which is correct so far as it goes but a complete list for example suppose three factors are taken into account it is not enough that these should be in fact vera causa there must be no other significant factor if there is a further factor not taken account of then the method is not able to discover the relative quantitative importance of the first three if so this means that the method is only applicable where the economist is able to provide beforehand a correct and indubitably complete analysis of the significant factors the method is one neither of discovery nor of criticism it is a means of giving quantitative precision to what in qualitative terms we know already as the result of a complete theoretical analysis john maynard keynes regarding the use of control variables one must also not overlook an important aspect that is seldom touched upon by those who use the statistical methods in question the variables included in the studies are treated as if the relationships between them in the population are random but variables can in fact have the values they have precisely because they give rise to the consequences they have the outcome thus to some extent determines why the independent variables have the values they do the randomised independent variables turn out in reality to be something other than what they are assumed to be and therefore also make it impossible for observational studies and quasi experiments to be anywhere near real experiments things often look the way they do for a reason sometimes the reasons are precisely the consequences that rules institutions and other factors are anticipated to give rise to what is perceived as exogenous is in fact not at all exogenous those variables that have been left outside of the causal system may not actually operate as assumed they may produce effects that are nonrandom and that may become confounded with those of the variables directly under consideration hubert blalock what conclusion do we draw from all this then causality is difficult and we should despite the criticism certainly not throw the baby out with the bathwater but adopting a healthy scepticism and caution when it comes to judging and evaluating the ability of statistical and econometric methods whether concerning causal graph theory or more traditional regression analysis to truly establish causal relationships is definitely to be recommended share on facebook opens in new window facebook share on x opens in new window x share on bluesky opens in new window bluesky econometrics a second best explanatory practice 13 feb 2026 at 10 56 posted in statistics econometrics comments off on econometrics a second best explanatory practice taken as a measure of causal explanatory power r squared does not fare any better the problem of explaining variances rather than levels shows up here as well if it measures causal influence it has to be influences on variances but we often do not care about the causes of variance in economic variables but instead about the causes of levels of those variables about which it is silent similarly because the size of r squared varies with variance in the sample it can find a large effect in one sample and none in another for arbitrary noncausal reasons so while there may be some useful epistemic roles for r squared measuring explanatory power is not one of them harold kincaid modern econometrics is fundamentally based on assuming usually without any explicit justification that we can gain causal knowledge by considering independent variables that may have an impact on the variation of a dependent variable as argued by kincaid this is however far from self evident often the basic causes are constant forces that are not amenable to the kind of analysis econometrics supplies us with as stanley lieberson has it in making it count one can always say whether in a given empirical context a given variable or theory accounts for more variation than another but it is almost certain that the variation observed is not universal over time and place hence the use of such a criterion first requires a conclusion about the variation over time and place in the dependent variable if such an analysis is not forthcoming the theoretical conclusion is undermined by the absence of information moreover it is questionable whether one can draw much of a conclusion about causal forces from simple analysis of the observed variation to wit it is vital that one have an understanding or at least a working hypothesis about what is causing the event per se variation in the magnitude of the event will not provide the answer to that question stanley lieberson s critique of the problem of basic causes strikes at the very heart of how social science interprets statistical control the things we want to explain are usually not randomly assigned often they are determined by the very covariates that are controlled for when making causal inferences based on statistical or econometric methods we must ask ourselves whether the variables we are controlling for are genuine confounders or the basic causes themselves without answering this question we are not really identifying causal effects at all analysis of variation the foundation of all econometrics can never in itself reveal how these variations are brought about only when we can tie actions processes or structures to the statistical relations detected can we say that we are obtaining relevant explanations of causation overly enamoured of axiomatic deductive modelling mainstream economists especially tend to forget that accounting for causation how causes bring about their effects demands deep subject matter knowledge and acquaintance with intricate contexts and their fabric as keynes already argued in his book a treatise on probability statistics and econometrics should primarily be seen as a means of describing patterns of associations and correlations instruments we may use as suggestions of possible causal relations by forgetting this economists will continue to be stuck with a second best explanatory practice share on facebook opens in new window facebook share on x opens in new window x share on bluesky opens in new window bluesky why we all care about inequality 11 feb 2026 at 09 15 posted in economics comments off on why we all care about inequality consider the example given by martin feldstein 1999 of a group of well heeled economists gathered at a conference or subscribers to an economic journal if each of us the economists were given 1 000 inequality in the united states would go up each of us would be better off and no one would be worse off so what is wrong with this asks feldstein apparently nothing let us now modify his example just a bit let us suppose that feldstein s fairy gives me 20 000 and each of the other participants is variously given between 25 and 75 cents feldstein s previous conclusions still hold everybody s welfare should be greater yet the effects i dare to suggest would be quite different many of the participants might refuse to accept their quarter some might leave it in the room others throw it away in disgust many would comment unfavorably upon the fact that i have received for some unfathomable reason 20 000 while other much more worthy members of the group would have to do with less than 1 1000th of that amount most would speculate on the reasons that lie behind such extravagant largesse on the part of the fairy what does this story illustrate first that many most of those who would have received 25 cents would not merely not feel better as feldstein suggests they should but would rather feel worse they would feel worse off because their feeling of justice and propriety would have been hurt and it will have been hurt because people always compare themselves to what they hold to be their peers thus income they receive is not only a means to acquire more goods and services it is also a tangible recognition of how society values them it is a social expression of their own...